Research Flashcards
Fundamental
Determinism
Observe
Empiricism
Test it Out
Experimentation
Repeat
Replication
Keep it Simple
Parimoney
Skepticism
Philosophic Doubt
Measures: Count/ Frequency
Repeatability
Event Recording Examples
Rate, Celeration, Percentage, Trials to Criterion
Measure: Duration
Temporal Extent
Use of a Timer
Time Sampling (momentary, partial, whole, PLACHECK)
Percentage
Latency, interresponse time
Temporal Locus
Cannot observe behavior
Permanent product
What the BX looks like
Topographical Measures
Force
Magnitude
Fluency
Rate
For continuous BXs
Time Sampling
Out of the Whole
Percentage
Time between SD and BX
Latency
Time between BXs
IRT
Count
Frequency
How efficient an intervention is
Trials to Criterion
Timing a BX
Duration
A special form of recording; a measure of the number of responses or practice opportunities needed for a person to achieve a preestablished level of accuracy or proficiency
Trials to Criterion
Measure for shoe tying, toothbrushing, hand washing
TTC
Underestimates Bxs
Whole Interval
Overestimates Bxs
Partial Interval
Either underestimates or overestimates BXs
Momentary Time Sampling
You can observe the whole interval
You want to increase the target behavior
Interested in an estimate
Whole Interval
You can observe the whole interval
You want to decrease the target behavior
Interested in an estimate
Partial Interval
You are unable to observe the whole interval
You are observing multiple behaviors
Interested in an estimate
MTS
The degree to which two or more observers report the same values after measuring the same events. It is the most commonly used indicator of measurement quality in ABA therapy.
Inter-Observer Agreement (IOA)
When does IOA drift?
After 2 mins
Best measure for IOA
Shorter lengths for the best results; under one minute
IOA - Simple and crude
Smaller count / larger count *100
Total Count IOA (Event Recording)
IOA - Increased Stringency
Calculate the total count IOA for each interval, then find the average of that
Mean count-per interval (Event Recording)
IOA - Most stringent
# of intervals of 100% IOA/total intervals * 100
Exact count per interval (IOA)
IOA - Stringent
Number of trials agreement/total number of trials *100
Trial-By-Trial (DTT)
IOA- Simple and crude
Shorter duration/longer duration *100
Total Duration (Duration)
IOA - Most stringent
Calculate total duration IOA for each occurrent, then find average of that
Mean Duration per Occurrence (Duration)
IOA - Moderately stringent
Number of intervals agreed over total interval *100
Interval by interval (interval recording / time sampling)
Scored interval IOA
Most stringent for behaviors occurring at low rates
For only intervals in which either or both observer marked an occurrence: intervals in which observers agreed/total intervals
Scored interval IOA
Unscored Interval IOA
Most Stringent for behaviors occurring at high rates
For only intervals in which either or both observer marked nonoccurrence: intervals in which observers agreed/total intervals
Unscored Interval IOA
Behavioral assessment involves the use of
A variety of methods
Two types of variables are examined in a behavior analytic study. They are:
Dependent and independent
A Target behavior is:
A behavior selected for change
A method used by a behavior analyst to gather data during behavioral assessment is
Interview, observation, checklist
A behavior cusp is a
a behavior that has consequences beyond the change itself
One characteristic of a well written operational definition is that it is
observable
clear
complete
For which of the following target behaviors would duration not be an appropriate measure?
hand raising in a class of 20 students, to signal a desire for attention
When using frequency measures of behavior, the behavior must have a(n) “” beginning and ending.
Discrete
Wolf (1978) states there are three judgements of social validity.
-the social significance of the goals
-the social appropriateness of the procedures
-the social importance of the effects
Which of the following is not a common error that occurs during a record review?
not operationally defining the target behavior clearly and concisely
Which of the following descriptions matches that of a topographical operational definition?
a definition that specifies the physical details that each response must show in order to be measured
In prioritizing problem behaviors for assessment/ treatment, one must consider:
-whether the behavior poses a danger to the individual or others
-whether changing the behavior will produce reinforcement for significant others
-whether changing the behavior will produce a higher rate of reinforcement for the individual
Which of the following is not a measure of repeatability?
Latency
A clear, measurable operational definition allows the behavior analyst to adhere to which value of science?
Replication
Which of the following is a dimensional quality of behavior of temporal extent?
Duration
Combining observation time with count results in a measure of
Rate
Converting count to the correct response in the previous question is done in order to:
make the measures gathered more meaningful
Celeration is a measure which shows:
how rate of response changes over time
When measuring a free operant behavior, the best measure to use is:
Rate
Interresponse time is a measure of
the amount of time that passes between one occurrence of a behavior and a second occurrence
The physical form or shape of a behavior is referred to as:
topography
The force or intensity with which a response is emitted is referred to as:
Magnitude
Measurement has validity when it:
results in data that directly measures what it sets out to measure
The results of measurement have external validity when they can:
be generalized beyond the conditions of the experiment
Reliability refers to
the extent to which observers looking at the client record in a consistent fashion
The most common way of reporting IOA is:
Narrative description
The most stringent way to measure I0A for data gathered using event recording is:
exact count per interval IOA
There are numerous methods for calculating IOA. T or F?
True
The formula for calculating IOA by the total count IOA method is:
smaller count/ larger count X100
When collecting discrete trial data, the behavior analyst would use measure IOA
Trial by trial IOA
IOA is the most commonly used indicator of measurement quality in ABA. T or F?
True
Measuring IOA requires that:
two indepedent observers concurrently measure
The threat to validity that exists when a researcher makes a claim about the applicability of a intervention to a population for whom it has not yet been tested is:
Generality across subjects
If you are measuring the total amount of time a baby sleeps at night, the dimension of concern for measurement is:
Duration
Tamara can correctly define 50 flashcards per minute. If she graphs this data, what dimension of behavior is she graphing?
Rate/ minute
Reliability of a measure refers to its
Consistency
An example describing a threat to internal validity due to history includes:
Subjects in the experimental group of a study are exposed to an unplanned event, such as teacher absence, during the intervention.
The bottom, horizontal border of a graph is called the:
abscissa
A small symbol, like a circle, is used on a graph to designate the value of a single measure. We call this a
Data point
If we wanted to compare several behaviors of several different students on a single graph, the best type of graph to use would be a:
Bar graph
The letters typically used to name a reversal design are:
ABAB
Baseline phases serve two functions. They are:
Descriptive and predictive
The first phase in a reversal design is:
Baseline
The reversal design includes how many minimum phases?
3
When using a reversal design, replication of the effect is accomplished during
Second intervention
A rationale for using a reversal design is:
its a strong indicator of the existence of a functional relation
If an intervention is added after a baseline condition, then removed, and the rates of the target behavior do not return to the original baseline levels, we says that “” has occurred
Irreversibility
An example of a behavior for which a reversal design would be appropriate
Disruptive behavior
A gym teacher instructs her students to use proper form while completing exercises in a fitness program. An occurrence of a response includes demonstrating the correct form. Which of the following measures must be used for observation of this behavior?
Topography
Celeration is a measure which shows:
•how the rate of response changes over time
Measurement conducted in a way which all responses are observed during a designated period of time is referred to as:
Procedural measurement
An experiment that shows the changes in behavior are a function of the independent variable and not the result of extraneous influences is said to to have:
Internal validity
Rate is the preferred measure for situations where an individual has an opportunity to make a correct and incorrect response because:
rate shows the proportional change in accuracy
The likely (best) measure a person would choose if they were interested in knowing how long an individual would wear their seat belt in the car without removing it is:
Duration
Nancy, a BCBA, has been asked to explain to her staff why classroom data for specific behavioral interventions is imperative. In her explanation, she should note that:
collecting direct and frequent measurement verifies the legitimacy of treatment and helps to ensure its effectiveness
When measuring a free operant behavior, the best measure to use is:
Rate
One necessary facet of measuring a behavior via permanent product is: the permanent product can only be produced by the target behavior. Another aspect is:
• each occurrence of the target behavior must produce the same permanent product
When a. timing procedure is used to record an observation, the most stringent method of 10A would be:
Exact count per interval
Accuracy, reliability and validity are concepts that behavior analysts are concerned about because each of these are:
considered quality indicators of measurement
Kris is concerned about the consistency of measurement of target behaviors for three of her students. Consistency of measurement refers to:
Reliability
This type of validity refers to when there is a correlation between a measure at one point in time and a measure at a future point in time.
predictive validity
A particular feature of a behavior that can be measured by a person conducting direct observation is referred to as a (n):
Dimensional quality
Two independent observers are positioned to watch a student who engages in head hitting behavior during a 10-minute period. The first observer counts 5 occurrences and the second observer counts 10 occurrences. The total count IOA is:
50%
The multiple baseline design is the least frequently used single subject design in ABA
False
The multiple baseline design can be used:
Across settings, subjects, or behaviors
When using a MBL design, verification occurs:
in the second and any subsequent baseline conditions
When using an MBL across settings, the researcher:
adds an intervention on the same behavior and subject across different settings or conditions
When using a multiple baseline design, the researcher adds an independent variable
after steady state responding occurs during baseline
The multiple probe design is a variation of the MBL.
True
If a researcher is conducting a multiple baseline design across settings and mid-way through the study a new setting becomes available the researcher should:
use a delayed multiple baseline design
If a teacher uses a MBL across subjects and has chosen 3 subjects and gathered 3 baselines, then the intervention should be added with:
• the individual with the most stable baseline
How often and the extent to which multiple measures of behavior yield different outcomes refers to which component of visual analysis:
Variability
Which answer below adequately describes the rationale for visually displaying data on a graph?
-to allow the practitioner to evaluate progress in an on-going manner, to make more timely change to an intervention if necessary
-as an effective tool in the provision of feedback for the individual(s) involved in the intervention
•to encourage meaningful assessment regarding the significance of behavior change
• all answer choices are correct
The value of the dependent variable is found on which axis on a line graph?
ordinate
According to LeBlanc et al. (2015), if the problem behavior is not observable but produces a measurable, physical change in the environment, the suggested measurement procedure is:
Permanent product
We can use the mean line to designate which of the following?
Level in a phase
Which of the behaviors below would it not be useful to graph data for on a simple line graph?
total tacts the student has learned
The primary method for evaluating data in single-subject research is:
Visual inspection
Which of the following uses does not pertain to a bar graph?
Shows a change in the dependent variable over time
With a changing criterion design, behavior changes:
In a step like fashion
The unique feature of a CCD (changing criterion design) is:
subphases are included in the intervention phase
In a CCD, the subphases are delineated by:
Vertical dashed lines
The number of subphases in a CCD is:
2 minimum but variable number can be used
With a CCD, criterion shifts are made when:
stability is achieved and behavior corresponds with criterion
Bi-directional changes are created by:
making the criterion temporarily less stringent
The CCD is well-suited for situations where behavior is:
To be altered gradually
A multiple treatment design “”
begins with a baseline condition
Usually
When an experimenter wants to compare the effects of 2 different treatments on a single behavior, they should use which design?
MBL
Read this statement and identify the dependent variable.
“Bella receives praise when she identifies a solution to a measurement situation correctly.
Identifying a solution
Belmont report included three things
-informed consent
-risk benefit analysis
-selection of subjects
What is IRB?
Institutional Review Board
When is IRB not needed?
When it is not available to public or not involving human subjects?
Definition of level
mean or average
Terms associated with Level
high / low
Definition of Trend
Direction of data path
Terms associated with Trend
increasing, decreasing, flat
Definition of variability
Balance
Terms associated with Variability
Stable, highly stable, variable, slightly variable
When to use split middle line
To estimate the trend when too difficult to eyeball
When are data points considered stable? -assessing variability
when approximately 80% of values
are within + 25% of the median value in a given
condition (Lane and Gast 2004)
The numbers within 25% of the median are called the “stability envelope”
change in level between end of one phase and beginning of next
Discontinuous
no change in level between end of one phase and beginning of nex
Continuous
How to calculate overlapping data?
find shared range between 2 adjacent conditions by drawing a horizontal line. Count the number of
Overlapping points and divided by the total points in the adjacent Condition.
Reversal Design Strengths
Very concinving design
Weakness in Reversal Design
Social and ethical concerns
When to use Reversal Design
When withdrawing baseline is possible and ethical
Variations of Reversal Design
BAB, ABACABAC, NCR reversal
Where is prediction for Reversal design
Baseline 1
Where is verification for reversal design
baseline 2
Where is replication for reversal design?
treatment / intervention 2
Strengths of Alternating Treatments
can compare treatments with worrying about sequence effects and without withdrawing intervention, intervention begins immediately
Weaknesses of Alternating Treatments
may cause multiple treatment interference, unnatural, max of 4 interventions
When to use alternating treatments
when comparing effect of interventions and cannot withdraw intervention and worried about sequence effects
Variations of alternating treatments
with a control condition, with a baseline
Baseline logic for Alternating Treatments
Each successive point plays all three roles
Strengths for MBL
does not require withdrawal, useful with generalization, feasible, easy to understand
Weaknesses of MBL
may not allow scientific control, weaker
When to use MBL
when withdrawal not possible, and comparing across subjects/settings/behaviors
Variations for MBL
Multiple probe, delayed MBL
Baseline logic for MBL
Prediction: baselines
Verification: Subsequent baselines after intervention introduced to previous baseline
Replication: subsequent intervention phases
Strengths of Changing Criterion Design
Does not require withdrawal
Weakness of CCD
can only look at one behavior at a time, very specific purpose
When to use CCD
When interested in step wise modification
When not to use CCD
When shaping, or teaching a new skill
Baseline logic for CCD
Prediction: baseline
Verification: varying lengths of phases, returning to a previous criterion (“mini reversal”)
Replication: subsequent phases of intervention
What is pivotal Behavior?
A behavior that when learned, produces corresponding modifications or covariations in other untrained behaviors
What is behavioral cusp?
A behavior that has sudden and dramatic consequences that extend well beyond the change itself, because it exposes the person to new environments, reinforcers, contingencies, etc.
What is direct measure?
What is measured is the behavior of interest (more valid)
-frequency recording
-Analog notes
-ABC data
What is indirect measure?
What is measured is in some way different from the behavior of interest (less valid)
-BCP checklist
-Questionnaire
-Permanent product
Measuring all individual instances
Ex: Count, duration, IRT, latency
Continuous Measure
Some instances may not be detected
Ex: time sampling
Discontinuous Measure
A response that does not have a clear beginning and end
Ex: smiling, eye contact
Continuous Behavior
Clear beginning and end
Ex: A single slap, a single bark
Discontinuous Behavior
Explain bar graph
With bar charts, each column represents a group defined by a categorical variable.
Explain histogram
With histograms, each column represents a group defined by a quantitative variable.
Elements of a Quasi-Design
Not a true experiment, no manipulation of an independent variable
Can approximate a group design (with no randomization of groups) or a single subject design (as a basic A-B design)
Used when resources are not available to conduct a true experiment
Can allow you to determine a hypothesis, but NOT to determine a functional relation
Elements of Group Design
True experiment
Consists of a control group and an experimental group – one gets the independent variable and the other does not
Uses inferential statistics to determine if there is a statistical difference between control group and exp. Group
^^This is how they determine a functional relation
Not often used in behavior analysis
Elements of Single Subject Design
True experiment
Reversal, MBL, Alt Tx, CCD
Uses descriptive statistics (AKA visual analysis) to determine a functional relation
Use if you have resources to do more than a quasi-design and/or if you need to determine an absolute functional relation
What is Type 1 Error
False positive
Concluding the intervention produced an effect when it did not
What is Type 2 error
False negative
Concluding the intervention did not produce an effect when it did
When does internal validity occur?
If the IV had an effect; was the experiment done right?
Face Validity
a form a measurement validity. Does the appear to be measuring what we think it does? Are we using the right form of measurement?
History
Threat to internal validity.
Event that occur between DV measurements that influence results.
Ex: family crisis, change in job, hurricane
Maturation
Threat to interval validity
Any change over time, may result from processes within subject
Ex: growing stronger, smarter, tired, bored
Treatment Diffusion
Threat to internal validity
Intervention is inadvertently provided when it should not be
A design looking at praise- but praise is given during baseline phases
Extraneous Variable
Any aspect of the experiment that must be held constant
-The experimenter held the extraneous variables, like # of question on the worksheet, constant during the experiment.
Confound
Uncontrolled factor exerting an influence on a DV
-subjects illness was a confound to the DV of the experiment
Non-parametric
One condition has a treatment, and one does not
Studying whether time out is more effective than not using time out
Parametic
Looks at variations of treatment
Ex: 10 minutes vs 20 minutes vs 30 minutes of an intervention
Studying the effects of varying durations of time out
