Parenting final exam SG PDF

Preview

Summary

Final exam for semester study guide...

Description

1

A. MEASUREMENT A-01, 02, 03, 04, 05; 12: Measure Frequency, Rate, Duration, Latency, IRT; Continuous Measures DIMENSIONAL QUANTITY Latency Duration Frequency (count)

FUNDAMENTAL PROPERTY

DEFINITION

Temporal Locus

Amount of time between stimulus and response

Temporal Extent

Amount of time between beginning and end of response cycle Number of responses that occur

Repeatability

Rate Interresponse Time (IRT)

Temporal Locus and Repeatability Temporal Locus and Repeatability

Celeration

Temporal Locus and Repeatability

Number of responses over time Amount of time that passes between end of one response and beginning of the next (similar) response Change in any of the above dimensional quantities over time (measures acceleration or deceleration)

Continuous/Direct/Dimensional Measures: 1. Event Recording: measurement of frequency or rate; record each instance of behavior -BX: moderate rate of response, short duration, clear onset/offset of response cycle a. Free Operant Event Recording: responses can occur at any time b. Restricted Operant Event Recording: responses are dependent on a specific antecedent (SD); track # of responses and # of SDs presented 2. Duration Recording: measurement of length of response cycle -BX: long response cycle, unclear onset/offset a. Duration per session: total amount of time an individual engaged in behavior *ex. 20 min. of tantrum bx in 1 hour session (e.g. 4 tantrums, 5 min. each) b. Duration per occurrence: amount of time individual engaged in behavior within a response cycle; do NOT add them together *ex. 2 tantrums: tantrum1 (7 min.) and tantrum 2 (3 min.) 3. Latency Recording: measure of the length of time between antecedent and response -measure from either onset or offset of antecedent to ONSET of response 4. IRT Recording: measure of time between responses -measure from either onset to onset or offset to onset a. Estimation of IRT: estimate of how long has passed between responses *ex. aggression occurs 12 times per hour 60 minutes/12 = 5 minutes between responses b. Average IRT: average length of time between responses *ex. aggression occurs at 5:00, 5:30, 7:00, 8:00 180 minutes/3* = 60 minutes between responses * number of IRT measures taken

2

A-06, A-07, A-13: Measure Percent of Occurrence, Trials to Criterion, Discontinuous Measures Discontinuous/Indirect/Dimensionless Measures: 1. Percent Occurrence: percentage of number of occurrences of restricted operant -does not accurately represent skill if too few opportunities to respond arise! *ex. 6 responses/10 opportunities (SDs) times = 60% occurrence 2. Trials to Criterion: number of trials completed before mastery was reached *ex. brushed teeth 27 times before able to do it independently 3. Discrete Categorization (Coding): classifying responses into discrete categories *ex. each opportunity for client to speak is coded as: 1 = full physical prompt 2 = partial physical prompt 3 = no prompt 4. Partial Interval Recording (PIR): behavior is recorded as “occurred” if it occurs at all during the time interval -recorded as [occurrences/# of intervals] -usually overestimates* rate of behavior *exception: underestimates high rate behavior because of the pause between intervals to record data** ** known as record interval: specified amount of time you get to record behavior before the next interval -USE TO DECREASE RATES OF BEHAVIOR 5. Whole Interval Recording (WIR): behavior is recorded as “occurred” if it occurs for the ENTIRE time interval -recorded as [occurrences/# of intervals] -used for high frequency + long response cycle behavior -underestimates rate of behavior -USE TO INCREASE RATES OF BEHAVIOR 6. Momentary Time Sampling (MTS): bx recorded as “occurred” ONLY if it is occurring at the end of a time interval -recorded as [occurrences/# of intervals] -used for when continuous observation is not possible and behavior occurs frequently -can either underestimate or overestimate rate of behavior *ex. teacher looks over every 10 minutes to see if student is completing work (target behavior) a. PLACHECK: type momentary time sampling in which a group is observed at the end of the time interval; count how many people are engaged in the target response -also known as Planned Activity Check 7. Permanent Product: method of measuring occurrence of behavior when it cannot be directly observed, but leaves a permanent change on environment -does not provide a full/accurate description of behavior *ex. # of cans in recycling, pieces of gum stuck on desk, bruises on arm from SIB

3

A-08: Interobserver Agreement (IOA) Coefficient of agreement between 2+ observers about a behavior of interest -helps to determine competence of data collectors, detect observer drift, and validate procedures -IOA should be collected for at least 1/3 of observations; IOA should be >80% Measuring IOA 1. Total Count: (total occurrences A/total occurrences B) x 100 -most commonly used with continuous measures (7 kicks/9 kicks) x 100 = 77% IOA 2. Percent Agreement: # of agreed/total # of occurrences x 100 -most commonly used with discontinuous measures (1 agree/(1 agree + 3 disagree) x 100 = 25% IOA 3. Average IOA: calculate IOA % for each interval separately, then find average Interval 1: (2/3) x 100 = 66% Interval 2: (0/0) x 100 = 0% 66+0+50+100/4 sessions = 49% IOA Interval 3: (1/2) x 100 = 50% Interval 4: (4/4) x 100 = 100% 4. Scored Intervals: (# of agreed occurrences/# of disagreed) x 100 -if all agree there was no occurrence, then omit the interval Interval 1: disagree Interval 2: omit (1 agree/2 disagree) x 100 = 50% IOA Interval 3: disagree Interval 4: agree 5. Unscored Intervals: (# of agreed non-occurrences/# of disagreed) x 100 -if all agree there was an occurrence, then omit the interval Interval 1: disagree Interval 2: agree (1 agree/2 disagree) x 100 = 50% IOA Interval 3: disagree Interval 4: omit

A-09: Measurement Procedures 1. Continuous Measures: measure direct, dimensional quantities of behavior -more accurate than discontinuous, but harder to measure -IF GIVEN AN OPTION, ALWAYS USE CONTINUOUS MEASURES!!! 2. Discontinuous Measures: measure occurrence vs. non-occurrence of behavior over time -less accurate than continuous, but easier to measure -higher rates of inter-observer agreement than continuous measures

4

A-10: Equal-Interval Graphs Components and Equal-Interval Graphs: 1. x-axis/abscissa: passage of time or environmental condition; 4(x):3(y) ratio length 2. y-axis/ordinate: dependent variable (quantification of target behavior); response measure -decreasing bx: number range of y-axis should be 1.5 – 2 times higher than bx rate collect at baseline -increasing bx: number range of y-axis should extend well beyond the goal rate 3. data points: relationship between dependent variable and measurement period of time -do NOT collect data points (not even a “zero”) when no data was not collected 4. data path: relationship between the dependent variable and independent variable 5. phase change: major change in the IV (solid vertical line) *ex. new schedule of reinforcement, baseline, beginning ABA services, discontinuation of services, client changed schools, parents divorced, new medication, etc. 6. condition change: minor change in a parameter of the IV (dashed vertical line) *ex. adjusting fixed ratio schedule, new medicine dosage, returned from vacation, sick with a cold, etc. 7. figure legend: concise statement that provides sufficient info to interpret the graph A-11: Cumulative Records Automated device that creates a graph that measures rate of response; responses move a stylus -steeper slope = increased response rate; flat line = no responses -x-axis = time; y-axis = rate -used in EAB -invented by Skinner for operant chamber -“cumulative” because line will never decrease

A-14: Choice Measures Preference assessments and reinforcer assessments (see I-07)

B. EXPERIMENTAL DESIGN B-01: Determining if Services are Behavior Analytic 1. Applied: implementation based on EAB; change behavior of social significance 2. Behavioral: behavior must be observable/measurable, need direct observation 3. Analytic: identify functional relations between behavior and environmental events through systematic and controlled manipulation 4. Technological: procedures are identified, described, defined; allow for replication 5. Conceptually Systematic: procedures used are a part of the principles of behavior 6. Effective: behavior change must make an impact; use data to revise behavior plan if not effective, ABA is accountable for these changes 7. Generality: behavior changes achieved should be maintained, transfer to other

5 settings, and spread to other behaviors

B-03: Functional analysis (experimental analysis, analog process) Arranged conditions designed to simulate hypothetical circumstances that support problem behavior -involves systematically manipulating antecedents and consequences to determine function of behavior in controlled conditions -FA is a process, not a set of procedures Functional Analysis Conditions: 1. Alone Condition: leave client alone; tests if behavior is maintained by automatic SR+ -cannot measure directly because non-social consequences cannot be controlled -provide no stimuli and record occurrences of self-stimulation -not to be used with young children; unethical 2. Attention Condition: give attention when client engages in target behavior; tests if behavior is maintained by social SR+ (attention type) a. Tangible Condition: give tangible when client engages in target behavior; tests if behavior is maintained by social SR+ (tangible type) *problem: tangibles can increase a bx even if the true function is something else (causes false positives + iatrogenic effects) *select tangibles based on naturalistic observation 3. Demand Condition: demand a task and remove it when client engages in target behavior; tests if behavior is maintained by social SR- (escape/avoidance) -presenting demand is EO (increases value of own removal as reinforcer) 4. Play Condition: test/control condition; demand-free access to attention/activities/tangibles with no programmed consequences -we want no EOs in place to evoke the target behavior -non-contingent, time-based schedule of delivery of preferred stimuli -when target behavior occurs, it is ignored (no reinforcement) - used for comparison with the test conditions Types of Functional Analyses 1. Standard FA: each condition is run for 3 sessions (usually uses multielement design) 2. Brief FA: each condition is run for 1 session (usually uses reversal design) 3. Trial-Based FA: each condition is run for 2 minutes; measures occurrence, not rate of bx -used in environments where variables cannot be controlled (ex. schools) -control—test—control structure 4. Extended Duration FA: observe client until problem bx naturally occurs at least once -used for low rate behavior that is too harmful to purposely evoke -only one test condition is run each session

6 *need intensive staff support and time to implement; can be unethical to leave client under a test condition for extended amounts of time

5. Burst Contingent FA: session does not begin (i.e. test condition not put in place) until the behavior begins to occur -used for low rate behavior that is too harmful to purposely evoke -conduct 2 tests and 1 control condition each session *does not require much staff support or time; staff must be able to quickly transition to the client to begin FA; potential carry over effects may skew data 6. Latency FA: target behavior is only allowed to occur once during session -used for harmful behavior -minimizes frequency of occurrence of problem behavior by ending session as soon as behavior occurs 7. Precursor Behavior FA: precursors to problem behavior are tested in its place -used for harmful behavior -precursors: behavior in a functional response class that reliably precedes other bx *precursors can be identified through direct observation and caregiver reports -can reduce risk of damage by placing contingencies on precursor behavior before it becomes high risk behavior 8. Protective Equipment FA: session is conducted with client wearing protective equipment -used for harmful behavior -note that behavior can be different when protective equipment is in place: i. rate of bx can decrease with protective equipment (obscures results) ii. protective equipment can reveal masked functions within the attention condition (adding equipment can make client want attention) Ethical Considerations: 1. Establishing Risk: determined by physician/medical exam; before conducting an FA, ask… a. Do the potential benefits outweigh the risk? b. Are protective measures in place to manage risks? 2. Managing Risk: review your high risk cases frequently and terminate session immediately when criterion is met (i.e. when behavior is evoked as many times as necessary) 3. Decreasing Risk: interview caregiver about severity, block self-injurious behavior (SIB) directed towards organs, use protective equipment, terminate session if redness/bleeding occurs until examined by doctor Ambiguous FA Results: -results are undifferentiated (i.e. no clear difference between test and control) in ~5% of cases -ambiguous results are more common if behavior has multiple topographies (always graph each topography separately) or if behavior occurs at a low rate 1. Reasons for Undifferentiated Results a. Multiple Control: behavior may have multiple functions

7 -2+ conditions will be high rate b. Automatic Reinforcement: behavior may reinforce itself more than other any other form of reinforcement -all conditions will be low rate c. Discrimination Problem: client may not be able to discriminate between each test condition so behavior will not change accordingly d. Missing Idiosyncratic Variables: relevant variables were missed/not tested 2. Solutions to Undifferentiated results a. Modify the type of design used b. Limit access to reinforcers (create EOs) c. Use a fixed sequence of test conditions [alone  attention  play  demand] d. Use specific signal stimuli before starting each condition (helps to differentiate) e. Allow bx to subside before moving to next condition (reduces risk of carry over effect) f. Try to determine and use idiosyncratic stimuli *Idiosyncratic Events Questionnaire (IEQ): used to obtain info from caregivers about atypical (i.e. individual specific) variables relating to the problem bx g. Use low-preference toys during attention and alone conditions Other Considerations: 1. Time and Resources a. Is there ample time and resources available? -if time is limited, test single function and conduct Brief FA b. Do you have control over the environment? -if poor control, conduct Trial-Based FA c. Does staff have training to conduct FA? -train others to conduct FA; additional training is needed to do… *data interpretation *identifying idiosyncratic variables *assessing risk management *treatment development -if training is not an option, then get consultation, supervision, or make referrals 2. Considering Differences in Between Populations a. Young Children i. use caregivers as therapists to reduce stress caused by separation ii. modify the clinic with items from their home to ease stress iii. do not test the alone condition (unethical!) b. Autism Spectrum Disorders (ASD) i. bigger focus on automatic positive reinforcement (stimming) ii. bigger focus on idiosyncratic variables (repetitive behavior) iii. sensory considerations Advantages/Disadvantages 1. Advantages i. determine actual functional relations ii. reveals methods to decrease problem behavior (i.e. potential treatment effects)

8

2. Disadvantages i. can only find what you actually test for (i.e. no functional relations can be determined if you didn’t set up relevant test conditions) ii. potential for iatrogenic effects (i.e. accidentally teach inappropriate behavior) -tangible condition most susceptible iii. complex/time/labor intensive

B-04: Withdrawal and Reversal Design A-B Design: baseline (A) followed by treatment (B) phase -supports only weak conclusions; poor internal validity -only used when other designs cannot be used; last resort! -forms the basis for all common single-case design Withdrawal/Reversal Design: requires minimum of 3 phase changes 1. Withdrawal Design: BL  IV  BL  IV -most straightforward design -strongest at establishing internal validity out of any design type -can begin with baseline or IV 2. Reversal Design: BL  IV1  IV2  IV1

3. Advantages/Disadvantages of Reversal/Withdrawal Design a. Advantages -most powerful method of demonstrating functional relations b. Disadvantages: -reversal (removing treatment) can be unethical -slow to complete; need to way for stability before changing phases -cannot be used for non-reversible behaviors (e.g. skill acquisition) -for reversal design, multiple treatments (IV1, IV2, IV3, etc…) can lead to confusion about which IV/combo was effective

9

B-05: Alternating Treatment/Multielement Design Multielement Design/Alternating Treatments: rapid alternation of two or more IV’s -all IV’s and baseline are implemented simultaneously -IV control is demonstrated when data path for one IV is VERY different from the others  target behavior: screaming -baseline: rate of screaming without punisher -IV1 (reprimand): rate of screaming decreases -IV2 (toy removed): rate of screaming greatly decreases to near zero rate when negative *IV2 shows most control over bx

Multi-Element with No Baseline

1. Multi-Element with No Baseline: baseline (no treatment phase) is implemented simultaneously with the other IVs 2. Multi-Element with Pretreatment Baseline: distinct baseline phase prior to implementing IVs 3. Multi-Element Plus Baseline and Final Treatment Phase: same as above, but with a final treatment phase after alternation in which each IV is implemented in isolation Considerations with Multi-Element Designs 1. Discriminability: ability to discern between two different conditions (IV’s) a. Mixed Schedule: condition (IV) changes without any signal b. Multiple Schedule: condition (IV) changes with a signal (SD) -use multiple schedules with multielement designs c. Other methods to enhance discriminability: reduce # of IV’s in place, include unique signal stimuli for each condition, provide instructions prior to phase change 2. Counterbalancing: intentionally varying non-essential environmental situations (settings, therapists, time of day, etc.) to minimize confounding factors (i.e. ensures changes in DV aren’t due to extraneous variables) 3. Strict Alternation: ensuring each IV is implemented in a certain order and each receives an equal amount of exposure (increases risk of sequence effect)  A-B-C-A-B-C (bad!) -do NOT do this with multielement design; keep it random!  A-C-B-A-A-B (good!) -restrictions can be used (e.g. “no more than two of the same condition in a row”)

10

B-06: Changing Criterion Design Uses sub-phases to increase/decrease your target bx in increments to reach a specific rate of response; IV control is demonstrated when rate of bx conforms to the criteria level; NO HIGHER OR LOWER! 1. Sub-Phase: a segment of a treatment phase; each sub-phase has a specific criteria rate for mastery; must be mastered before new sub-phase can be implemented -each sub-phase acts as a baseline for the next sub-phase

2. Strengthening Internal Validity of Multi-Element Design: a. have at least 3 criteria goals b. vary the length of time the client is in each sub-phase c. use bi-directionality: changing criteria to a previous goal -DV should conform to the new (previous) criteria if there is

3. Considerations a. keep criteria changes small so client can feel successful -if too small, you may not be able to detect behavior change b. stable behavior: large increment changes; variable behavior: small increment changes c. only used for behavior that’s already in client’s repertoire 4. Advantages/Disadvantages a. Advantages i. tr...