Ch 7 Analyzing Behavior Change: Basic Assumptions & Strategies Flashcards
foundation of single-case experimental designs. provides framework for using patterns of bx in the baseline phase to demonstrate experimental control and establish functional relationship between IV and DV.
3 types- prediction, verification, replication
baseline logic
phase in intervention where target behavior is measured without any intervention or treatment in place. provides a starting point to compare future behavior changes and helps determine whether an intervention is effective
-conducted before the IV is introduced
-behavior is measured under natural/typical conditions
-helps establish a pattern or trend of responding
-used to make predictions and assess experimental control
-can show level, trend, variability
baseline
demonstrating that behavior change does not occur without the intervention; process of confirming that a functional relationship exists between a change in the independent variable (e.g. an intervention or treatment) and a change in the dependent variable (e.g. behavior). It’s one of three components of baseline logic used used in single-subject experimental designs (prediction, _____, replication).
-supports internal validity
-confirms that the intervention caused the behavior change
verification
baseline phase in which the target behavior fluctuates unpredictably- there is no clear trend, level, or stability. this makes it difficult to predict future behavior and can weaken experimental control if not addressed.
-behavior shows inconsistent patterns
-makes prediction and interpretation of treatment effects harder
-may result from environmental variables, lack of stimulus control, or measurement issues
-important to extend baseline until stable data are observed
e.g.
day 1 20%, day 2 60%, day 3 10%, day 4 40%, day 5 15%= no clear trend, level, or stability
variable baseline
method used in experimental analysis and single-subject designs where the IV (independent variable; e.g. an intervention) is introduced only after the DV (dependent variable; behavior) shows a stable pattern under baseline conditions. ensures changes in behavior can be confidently attributed to the intervention
-focuses on achieving stable baseline data before introducting treatment
-ensures there’s a clear prediction of future behavior based on baseline
-strengthens internal validity and experimental control
-often used in reversal (ABAB) and multiple baseline designs
-minimal variability, clear trend, predictable pattern, usually determined by visual analysis across 3-5 data points
e.g. tantrum frequency;
day 1- 4, day 2- 4, day 3- 3, day 4- 4 since this is __, the intervention can now be introduced. any change in behavior can be more confidently attributed to the treatment.
steady state strategy
condition in which a behavior shows a consistent pattern over time- with minimal variability in rate, frequency, or intensity- under a given set of environmental conditions (e.g. baseline or intervention). it indicates that the behavior has reaches a stable level, making it possible to detect changes when a new condition is introduced.
-behavior occurring in a predictable pattern
-low variability, clear trend, or consistent level
-suggests experimental control is possible
-often used to decide when to introduce or change phases in single-subject designs
- ensures a valid baseline or intervention phase and helps differentiate real treatment effects from natural variability
-essential for SSED (single-subject experimental designs)
e.g. hand raising frequency at baseline
session 1- 2 times, session 2- 2 times, session 3- 3 times, session 2- 2 times= ___- it’s stable, predictable, and ready for the introduction of the independent variable
steady state responding
period during which the target behavior occurs at a consistent level or trend before an intervention is introduced. in SSEDs, a _____ is critical for making valid predictions and assessing whether a functional relationship exists between the IV (intervention) and DV (behavior)
-behavior shows little to no variability, clear trend (increasing, decreasing, or flat), predictable pattern
-suggests that external variables are not influencing the behavior
e.g. out of seat behavior
day 1-4, day 2- 4, day 3- 5, day 4- 4= a ____, making it a good point to introduce a behavior reduction intervention (e.g. differential reinforcement)
stable baseline
experimental methods used in ABA to evaluate the effect of an intervention on the behavior of one individual (or a small group), where the individual serves as their own control. behavior is measured repeatedly over time during baseline and intervention phases to demonstrate a functional relationship between the independent and dependent variables.
-focus on individual behavior change
-repeated measurement
-includes at least one baseline phase (A) and one intervention phase (B)
-uses visual analysis of graphed data
-seeks to demonstrate experimental control (functional relationship)
single case design
types
AB
reversal (ABA or ABAB)
multiple baseline
changing criterion
alternating treatments (ATD)
withdrawal
process of repeating an intervention or experimental condition to determine whether the same behavior change occurs again. critical component of baseline logic and is used to demonstrate reliability and validity of a functional relationship between the IV (intervention) and DV (behavior)
replication
3 types
direct
systematic
internal/within experiment
a clearly defined question that guides the purpose of an intervention or experimental study. identifies behavior of interest, population, and IV (intervention), and often setting or context. helps determine what the practitioner or researcher is trying to discover, change, or evaluate.
-focuses on specific observable bx
-clearly defines IV (intervention or treatment)
-states population or individual being studied
-can be measured and answered using data and analysis
e.g. does use of a token economy increase on-task bx in a special education classroom?
does a DRA decrease tantrums in a preschool child with ASD?
research question
anticipating future level of behavior based on baseline data before an intervention is introduced. 1 of 3 components of baseline logic (along with verification and replication) used in single-case experimental designs to demonstrate experimental control.
-based on trend, level, and variability of baseline data
-used to determine what bx would look like without intervention
prediction
improvements in performance that occur due to repeated exposure to a task or behavior- NOT because of an intervention. they can confound experimental results by making it seem like bx changed because of treatment, when in reality, it changed due to familiarity, repetition, or learning the task over time.
-type of confounding variable that can threaten internal validity
practice effects
type of experimental procedure to examine effects of different values (levels or intensities) of the IV on bx. helps determine how much of the intervention is needed for a bx change and whether more or less of it produces better, worse, or no effects.
-involves sytematic variation of the magnitud, duration, or rate of the IV to help determine optimal dose of an intervention
-used to fine-tune tx effectiveness
-different from component analysis, which tests whether each part of a tx is necessary
e.g. Duration of access to a reinforcer (e.g., 2 min vs. 5 min of tablet time)
Delay before reinforcement is delivered (immediate vs. delayed)
Intensity of a prompt (partial vs. full physical)
Dosage of treatment (e.g., 1 vs. 3 sessions per week)
parametric analysis
degree to which we can confidently say that a change in the DV (bx) was caused by the IV (intervention) and NOT by other uncontrolled variables. demonstrates a functional relationship between bx and tx.
internal validity
the intervention, treatment or manipulated condition the BCBA intentionally changes to see its effect on behavior. it’s what you do to the environment to influence behavior; the “cause” in the cause effect relationship
independent variable
extent to which the results of an intervention or experiment can be generalized to other people, settings, behaviors, or times beyond the specific conditions of the study.
-focuses on generalization of bx change
-important for real-world application of interventions
-strengthened through systematic replication across people, settings, and behaviors
external validity
structured plan used to test whether a functional relationship exists bewteen an IV (intervention) and DV (behavior). involves systematic manipulation of variables, repeated measurement, and baseline logic (prediction, verification, replication) to establish experimental control.
-involves manipulating IV
-repeated measurement of target bx
-uses baseline and intervention phases
-demonstrates functional relationships
-ensures internal validity
experimental design
achieved when the IV (intervention) is shown to be the only cause of a change in the DV (behavior). it means the bx reliably changes when the intervention is introduced and returns or stays the same when it is removed or withheld, demonstrating a functional relationship.
-shows the IV alone is responsible for bx change
-demonstrated through baseline logic
-rewuires repeated, systematic manipulation of variables
-core goal of single-case experimental designs
experimental control
pattern in which target behavior is decreasing over time before any intervention is introduced. this trend can make it difficult to interpret whether a laterbx change was due to the intervention or whether it was naturally improving
-may suggest bx is improving without intervention
descending baseline
the target behavior being observed and measured to assess whether an intervention is effective. it’s the outcome of interest and basis for making data-driven tx decisions.
e.g. of ___
number of aggressive bxs per day
duration of tantrums during a task
latency to respond to a prompt
rate of manding
percentage of correct responses
dependent variable
uncontrolled factor that co-occurs with the IV and may influence the DV, making it difficult to determine whether the intervention actually caused the bx change. it muddles the interpretation of the results by introducing alternative explanations for what caused the bx change
-actual threat to internal validity
-affects DV and overlaps with IV
-occurs alongside intervention and affects the bx being measured
e.g. medication changes during an aba program targetting aggression
a new therapist starts during the intervention phase
client begins getting more sleep at night during the bx plan
parent starts providing attention for appropriate bx, but this wasn’t part of the intervention
confounding variable
any variable other than the IV that might influence the DV. these variables are not intentionally part of the experiment but can confound the results of an intervention by making it unclear whether the intervention or something else caused the behavior change.
-potential threat to internal validity
-might affect DV
-may be environmental, biological, or social. should be controlled, monitored, or ruled out
e.g. noise levels in the room during sessions
lighting or temperature change
fatigue or hunger
presence of different therapists
unplanned schedule changes
extraneous variable
pattern in which target behavior is increasing over time before any intervention is introduced. this upward trend can make it difficult to interpret whether a future increase (or decrease) in bx is due to the intervention or natural trend
-bx is already increasing in baseline
-may suggest the bx is worsening naturally or improving without intervention
ascending baseline
logical fallacy that occurs when someone assumes that because a specific outcome occurred, the intervention must have caused it- without ruling out other possible causes. a flawed form of reasoning that can weaken experimental conclusions if not addressed.
logical form (fallacy)
if A, then B. B occurred, therefore A caused B. this is not logically valid without further verification.
-what if another variable caused it? (e.g. new teacher, scheduling, weather, sleep, medication change)
affirmation of the consequent
