1
Q
what is assessment?
A
- the systematic measurement of the impact an intervention has on someone
- highly important in any intervention where we want to see change
- can tell us if an intervention is having the intended effects
2
Q
what does assessment not show?
A
- what caused the change
- this is extremely important if we want to apply an intervention to other people beyond one client
3
Q
what is a research design?
A
how are the conditions arranged so that one can make assumptions on how the intervention was responsible for the changes.
4
Q
where is research design mostly done?
A
- experimental group research
- usually done with a group that receives no treatment, a placebo group, and a group that receives regular treatment
5
Q
what is between-group research?
A
people are randomly placed into groups and are measured before and after the intervention to see if there are similarities or differences.
6
Q
where are applied interventions used?
A
single-case experimental designs.
7
Q
what is data evaluation?
A
- looking at the data and seeing if the changes made during the intervention were successful
- is the data different across different situations
8
Q
what is statistical analysis used for?
A
- seeing changes that might be hard to see with our eyes, or changes that appear to be small
- analyzed based on a bunch of factors, like probability level, to see if the differences are due to the intervention
9
Q
why is it important to see if a behavior change can be extended to other areas of a client’s life?
A
- to help their functioning as much as possible
- knowing what caused the change can help us maximize the benefits
10
Q
what is a single-case experimental design?
A
- the arrangements used to draw inferences about the impact of an intervention
- usually observes client several days before intervention
11
Q
why are single-case experimental designs functional?
A
- they demonstrate what can happen when altering the experimental condition
- does it change behavior
12
Q
why are single-case experimental designs individualized?
A
it tailors the intervention to that person.
13
Q
what the most important requirement of single-case experimental designs?
A
- repeated observations of performance overtime.
- client’s performance is observed many times (usually daily) before, during, and after the intervention
- allows observers to examine performance and how it changed with the intervention
14
Q
what is the baseline phase?
A
- the initial observation before the intervention.
- it can tell us about the level of behavior beforehand
15
Q
what is a descriptive baseline function?
A
describes the rate of the client’s behavior beforehand.
16
Q
what is a predictive baseline function?
A
predicts future behavior if the intervention is not implemented.
17
Q
how can we tell if performance is stabilized?
A
characterized by the absence of a trend (or slope) in the data and little variability in performance.
18
Q
why is it important for behavior to be stabilized in the baseline phase?
A
so we can later see if there was an impact caused by the intervention.
19
Q
what are trends in data?
A
the tendency for performance to increase or decrease consistently over time.
20
Q
why are trends problematic?
A
if a behavior that we want to decrease is already decreasing in the baseline phase it can make it harder to tell if the intervention was effective.
21
Q
what is the most ideal baseline?
A
a stable one with little to no change in behavior.
22
Q
how do we know if an intervention will last?
A
- has good stimulus control
- someone is learning a new behavior that is reinforced so it becomes easier and is maintained with weaker reinforcers
23
Q
what is variability in data?
A
the fluctuation or stability in the subject’s performance overtime.
24
Q
why is variability in data problematic?
A
it can make it difficult to draw conclusions if an intervention worked.
25
what is ABAB/reversal designs?
- family of experimental arrangements in which observations of performance are made overtime about a client
- examines the effects of an intervention by altering between the baseline condition (A phase), when no intervention is in effect, and with the intervention condition (B phase)
- this is repeated again to complete four phases (hence ABAB)
26
what do ABAB/reversal designs always begin with?
- the baseline phase to see if pre rate behavior will remain stable
- helps us to observe current behavior and predict future behavior if intervention is not there
- should only move forward if behavior is stable
27
what happens during the intervention phase in an ABAB/reversal design?
we test to see if there will be a difference from the predicted behavior and the predicted behavior after the intervention.
28
what happens in the third phase (second A) in an ABAB/reversal design?
- intervention is withdrawn and baseline conditions are restored
- describes current performance and predicts what future behavior will look like if it continues
- asks if behavior will still look the same with the first intervention having taken place
29
what happens in the final phase (second B) in an ABAB/reversal design?
- it describes performance, testing if it departs from the previous phase behavior (the pre-rate), and if it is the same as the previous intervention phase.
30
what is an AB design?
- one baseline and one intervention
- not usually considered to be a configuration as it does not provide a description, prediction, or test of predictions
31
what are AB designs associated with?
level changes, transitional phases, or both.
32
why are AB designs problematic?
- we are not sure if the change is due to other factors
- possible regression towards the mean (extreme behavior that is more likely to become average overtime because the extreme behavior is not ordinary)
33
what is an ABA/BAB design?
- baseline, intervention, baseline (ABA)
- intervention, baseline, intervention (BAB)
- there are at least two phases of the same to show a return to something, whether that be the baseline or intervention
34
why is the ABA/BAB design problematic?
- ABA can be a problem as you leave the person in the baseline
- BAB can be difficult if you leave the participant in the treatment condition, which is better but sometimes the agency might not like the treatment
35
how many phases tend to show a stronger basis for drawing conclusions about an intervention?
four or more.
36
what is an AB(1) B(2)AB(2) design?
- has more than one type of intervention
- B1 is the intervention that did not work
- B2 is the intervention that did work
37
what is non-contingent delivery of a reinforcing consequence?
- another reversal phase possibility
- delivering a consequence that is not based on a certain behavior (ex: giving treats to a dog even if they did not do anything)
38
what is an ABCB design?
allows one to compare treatment between baseline, treatment B, and a control condition which could be non-contingent reinforcement.
39
what is an ABACABAC?
used to compare two treatments.
40
what is an ABACA(B+C)ABAC design?
- compares two treatments applied separately and two treatments applied together
- it is easy to overwork this design if you introduce too many treatments
41
what is data overlap?
- when data points appear in the same place
- we do not want this
- if happening, control the relevant variable to reduce variability
42
how do you strengthen an ABAB/reversal design?
- control variables
- add more reserverals/replications
- increase intensity of treatment
- collect more data per data point
- add stimuli to aid in the discrimination of conditions
43
what is helpful about an ABAB/reversal design?
- good for allowing us to see the impact of intervention
- not necessary to randomly assign people into groups
- do not need a lot of people
- you do not need an untreated control group
44
what is limiting about an ABAB/reversal design?
- can be difficult to know if the intervention will be withdrawn depending on the person giving it (ex: teachers, parents)
- sometimes behavior changes cannot be removed as simply withdrawing the intervention
- possible ethical questions regarding taking back treatment (ABA design)
- the longer a treatment is in place, the harder it is to get it to reverse back to baseline. stimulus control becomes stronger and other reinforcers may develop that will maintain the behavior
- the problems that can arise from ABAB can make it undesirable in applied situations.
45
what was the azrin study?
- looked at annoying habits to tics
- treated many people, this is known by comparing baseline and intervention data
- this study was easily replicated
46
what are multiple baseline designs?
- effects are demonstrated by introducing the intervention to different baselines (behaviors or persons) at different points in time
- if each baseline changes with the intervention, we can say that the intervention caused it
- interventions do not need to be redrawn
47
what is the multiple baseline design across behaviors?
- looks at multiple behaviors in a person
- baseline data is gathered on two or more of these behaviors
- after baseline behavior is stable for each behavior, intervention is applied to one of them. if effective, one would expect to see a change in the behavior that is receiving the intervention
- then it is applied to another behavior, and so on
48
what are traits of a multiple baseline design across individuals?
- most common and most reliable
- behaviors do not need to be independent
49
what are traits of a multiple baseline design across behaviors?
- behaviors must be independent
- try to select an unrelated baseline
50
what are traits of a multiple baseline design across situations?
- stimulus control must not generalize to other situations
- maximum differences work best
51
what are the strengths of a multiple baseline design?
- does not require reversal
- not affected by irreversibility
- controls the amount of time on the baseline without the independent variable
- if an intervention is not helpful, it can be fixed before it is used in other situations, behaviors, people, etc.
52
what is the minimum amount of baselines in a multiple baseline design?
- two baselines
- three or more is better
53
what are the limitations of a multiple baseline design?
- in designs with situations, settings, or time periods, it is possible that altering the behavior in one situation may have it spread to others
- does not happen often
54
what is changing criterion design?
- shows the effect of the intervention by showing that behavior matches certain criteria for performance that is set for either reinforcement or punishment
- starts with a baseline phase in which multiple observations about a person are made
- after the baseline phase (A), the intervention (B) begins
- uses of several sub-phases (b1, b2, to bn)
55
what happens if a person meets a certain criterion in a changing criterion design?
- if a person meets it, they will be rewarded
- if a person consistently reaches these criteria for several days, then it will be upped slightly. this repeats until the desired level of performance is met
56
why may a mini reversal be used in changing criterion design?
adding a phase in which behavior changes in the opposite direction to follow a criterion reduces the ambiguity about the influence of the intervention.
57
what are the strengths of a changing criterion design?
- best for settings where behavior learning is gradual
- allows us to see if an intervention works and if it doesn’t, we can change it so it can
- since these designs focus on a very few people, it can make it easier to study things that are rarer (like a particular genetic disorder)
58
what is problematic about large changes of behavior in a changing criterion design?
- the specific effect of the intervention in altering behavior will not be clear
- this is the reason that a mini-reversal phase (return to a prior criterion level but not back to baseline) is sometimes implemented
59
what are true experiments?
- experiments (like ABAC, multiple baseline designs, changing criterion designs) that are rigorous to see if the intervention impacts behavior
- may not be difficult to apply as we want to optimize change (goes against ABAB) or it may make it difficult to conduct the design (multiple baselines)
60
what are basic features of a group design?
requires at least two groups: one receives the intervention (the experimental group) and the other does not (the control group) to determine whether the intervention is effective, some outcome (e.g., rate of the target behavior, level of depression or anxiety) in the experimental and control groups is compared.
61
how is assessment used in group designs?
usually made before and after the intervention is delivered.
62
what do most group designs rely on?
a few observations of many individuals (subjects).
63
why is random assignment important in group designs?
if clients are not randomly assigned to groups, the likelihood is greater that the groups may be different in their performance of the target behavior before the program even begins or differ at the end of the program for some reason unrelated to the intervention.
64
what are randomly controlled trials?
when groups designs are used to study interventions (e.g., in education, medicine, psychology) and clients are assigned randomly.
65
what are combined designs?
when both group and single-case designs are used.
66
what is evaluating data?
- describing and making inferences about the changes seen an intervention
- focuses on whether the change is likely to be reliable rather than due to chance fluctuations in performance
67
how are most group designs evaluated?
statistically.
68
how are most single-case designs evaluated?
comparing performance during baseline with performance during the experimental phase.
69
what is visual inspection?
reaching a judgment about the reliability or consistency of intervention effects by visually examining the graphed data.
70
what are the changes in the mean/average?
shows a change from phase to phase in an expected direction.
71
what are the changes in a trend/slope?
- shows a change from phase to phase
- ex: showing no change in slope in the baseline and accelerating slope in an intervention
72
what are shifts in level?
shows a change from phase to phase where behavior from the first day of a new phase changes from the last day of phase one.
73
what is latency of change?
- period of time between onset or termination of one phase or condition (ex: baseline to intervention) and changes in performance
- closer the time the behavior change occurs after the conditions have been altered, easier to say that the change happened because of the intervention
74
what is non-overlapping data?
the pattern in which the values of the data points during the baseline phase do not approach any of the values of the data points obtained during the intervention phase.
75
what research design is very important for graphing data
single-case experimental designs.
76
what are simple line graphs?
- the most commonly used method of plotting data in single-case designs that consists of noting the day-by-day (or session-by-session) performance of the client over time
- data points within each phase are connected to produce a line
77
what are the strengths of a simple line graphs?
- easy to determine how something is performing at a glance
- changes in mean, level, slope, and the rapidity of changes in performance are especially easy to examine
78
what are cumulative graphs?
plots performance in an additive fashion over time.
79
what are bar graphs?
- presents the data in vertical or occasionally horizontal columns to represent performance under different conditions
- each bar or column represents the mean or average level of performance for a separate phase
80
what is the problem with bar graphs?
can be very misleading.
81
what is a multi-element design?
- revved up version of the reversal design
- both treatments are given each day in a random
order
- pairing an instruction or a clear discriminative
stimulus with each of the two conditions also
helps to magnify the effect of the design
- it is best to separate the two treatments with a
baseline phase when possible
82
what is an alternating treatment design?
varies it across days but it does not control variables that may change on a day- to-day basis such as motivating and abolishing operations.
83
what are carryover effects?
exposure to one treatment can influence performance on a second treatment.
84
what are generalization effects?
- if one person administers both treatments, personal
stimulus control can produce generalization from the effective to the less effective
treatment
- using different stimuli, contexts, in both situations, can minimize this effect
- if the person has learned a new technique they may engage in it in both settings
behavioral analysis
flashcards
Decks in class (14)
# Cards
-
introduction65
-
principles of operant conditioning47
-
assess behavior49
-
quality of assessments23
-
evaluating interventions84
-
functional behavioral assessment51
-
positive and negative reinforcers61
-
extinction39
-
technique variations77
-
generalization53
-
self-control47
-
clinical applications91
-
social, ethical, and legal contexts57
-
quiz questions50
