Sped 503A Lesson 4

Front 1

Why are Formal Methods
of Observation Needed?

Back 1

Unsystematic Observations May Be
Inaccurate Because:
•  Unrepresentative
•  Have imprecise definitions of
behavior
•  Subjective

Front 2

Stages of Behavioral
Assessment

Back 2

-Screening and General Disposition
-Definition of Behavior and Desired Outcome
-Pinpointing Target Behavior
-Monitoring Progress
-Follow-up

Front 3

Stage 1 of Behavioral Assessment: Screening

Back 3

- Purpose- does the problem warrant a program?
- Methods used:
Interview people involved
Medical exam (if needed)
Ecological assessments
Observation (questions to answer)
Summarize: preferred activities, sensory, strengths, inappropriate behavior, what they like, how do they spend their time

Front 4

Stage 2: Definition of
Problem

Back 4

  What are the general behaviors
that need to be addressed?
  Steps 1 and 2 are often combined

Front 5

Stage 3: Pinpointing
Target Behavior

Back 5

Defining behavior:
  Behavior =
  Dead man’s test

Front 6

Stage 3: Pinpointing
Target Behavior
Three characteristics of a good definition (Hawkins and Dobes)

Back 6

1) Objective, observable characteristics
2) Readable (unambiguous- strangers test)
3) Points out what not the beahvior as well as what is

Front 7

Questions to help pinpoint behaviors

Back 7

-what does the student do?
-what do you wnt the student to do?
-cn you count the number of times or how long behvior occurs?
-Can my definition pass the stranger test?

Front 8

Definint Target Behaviors
Constructs vs Behaviors

Back 8

Construct (Broad description, how we interpret event, subjective, what a student is, hard to measure, usually and adjective)
Behavior (narriow description, describes event, objective, what student does, easy to measure, usually a verb)

Front 9

Two types of behavior

Back 9

Function-based
Topographically based

Front 10

Function-Based

Back 10

focus is on a response class (group of behavior that serve the same purpose/function)

Front 11

Topographically based

Back 11

Identifies behavior based on form (how it looks)
ex: hand raising, golf swing, writting a letter
use when behavior does not result in the same outcome each time

Front 12

Why is function based preferred?

Back 12

- don't have to define every behavior
ex: contact with body part in mouth for at least 3 seconds, initiates a beahvior requested by a teacher within 5 seconds (compliance)

Front 13

Questions to ask when choosing a behavior to change.
we know about:
1. changing socially significant
behaviors
2. objectively defining behavior

Back 13

Does the behavior:
  produce reinforcement in natural setting after program? (what will maintain behavior after intervention?)
  serve as a key prerequisite? (
  disrupt learning?
  provide access to other environments?
  cause social exclusion?
  prevent from obtaining reinforcers?
  serve as a cusp or pivotal behavior? (opens doors to new contingencies)

Front 14

Prioritizing which behavior to choose

Back 14

  Frequency of the problem
  Causing physical harm to self or
others?
  Will the behavior change help the
learner?
  Age appropriateness
  Is this really the behavior you want
to change?
  Are we replacing the behavior with
an alternative?
  Relevance of behavior rule
  “So what” test
  Potential for success (check research, past experience, resources available, can you control the reinforcer?)

Front 15

Selecting Replacement
Behaviors

Back 15

  Selecting behavior to change is more
than defining inappropriate behavior
  Trap of decreasing behaviors
  Replacing target behavior with
functional alternative
  Fair-pairs (replacing functional behavior with target behavior)
  Increasing _____ will decrease____.

Front 16

Two Key Terms for Selecting Target Behaviors

Back 16

Social Validity (other can notice change)
Habilitation (degree to which a person's repetoire or skills set maximizes and short and long term reinforcers for self and others)

Front 17

Set Goals

Back 17

What do we want to see (behavior)?
How can we see it (conditions)?
When are we there (criterion)?

Ex:
Increased social interactions (initiating with peers)
During recess and lunch
The mean number of interactions for tyical children is 2 per minute.

Front 18

Stage 4: Monitor Progress

Back 18

Data collection during intervention
Allows for fast decision making regarding program

Front 19

Stage 5: Follow Up

Back 19

Data collection after intervention
Did gains maintain and generalize

Front 20

Indirect methods for gathering information about behavior

Back 20

Interviews
Review of Records
Incident Reports
Tests and rating scales
Indirect observation

Front 21

Direct Methods for Gathering Information

Back 21

Anecdotal (ABC) recording
Permanent Products
Event Recording (Frequency Counts, Rate, %)
Interval Recording/Time Sampling (Partial I.R., Whole I.R., Momentar Time Sample)

Front 22

Guidelines for ABC recording

Back 22

Write down everything the student says, does, and everythign that happens.
Write down time behavior occurs
Observe over a period of a few days

Front 23

Ecological Assessments

Back 23

Assessment between a person and their enviornment, is person's skills set a good match with the environment

Front 24

Uses for an Ecological Assessment

Back 24

1) Physical aspects of enviornment, instructional varriables that are mismatched.
2) Looking at next area of functioning, write down skilsl required in next LRE and teach those skills for the next area of functioning

Front 25

Review, we know behavior has:

Back 25

Repeatability (how often)
Temporal Extend (how long)
Termporal Locus (when)

Front 26

Which dimension of behavior:
How often?

Back 26

Frequency

Front 27

Which dimension of behavior:
How long?

Back 27

Duration

Front 28

Which dimension of behavior
How long before start?

Back 28

Latency

Front 29

Which dimension of behavior
How strong?

Back 29

Magnitude

Front 30

Which dimension of behavior
How does the behavior look?

Back 30

Topography

Front 31

Define permanent products and advantages.

Back 31

Result of behavior.
advantages: outcomes agreed upon, can use video, can record behavior later.

Front 32

Event recording

Back 32

counting the number of times something happens

Front 33

steps in event recording

Back 33

1) Decide when you'll observe
2) Keep track for start and stop times
3) Count the number of times behavior occurs

Front 34

Advantages of Event Recording

Back 34

easy
Great when you want to increase or decrease number of times something occurs

Front 35

Event Recording considerations

Back 35

-discrete target behavior
-not occur at high rate
-not occur for long duration
-opportunities to respond should be constant (recorded in same time intervals, and number of opportunities)

Front 36

What ot do if opportunities to resond or observation time differ

Back 36

convert data

Front 37

Ways to convert data

Back 37

percentages
rate (frequency/time)

Front 38

When to use Frequency
What to Report
What it Tells.

Back 38

-use when time and opportunities are constant
-report number of targeted responses
-tells about number of times target behavior occurs

Front 39

When to use Rate
What to Report.
What it Tells.

Back 39

Use when time varies or when interested in fluency.
Report number of targeted responses within a time period.
Tells about fluency or proficiency

Front 40

When to use Precent.
What to Report.
What it Tells.

Back 40

Use when opportunities vary.
Report number of target responses within a specified number or responses.
Tells about accuracy.

Front 41

Steps for Interval Recording and Time Sampling

Back 41

1) Define target beahvior
2) Decide how long you will observe
3) Divide the observation session into equal intervals

Front 42

Interval and Time Sampling
-provides what?
-is useful when?

Back 42

Provided an estiate of behavior
Useful with continuous and high-rate behaviors

Front 43

Partial Inteval Recording

Back 43

Intervals is scored when any target response occurs within a given interval (did the behavior happen during the interval?)

Front 44

Whole Interval Recording

Back 44

Interval isscored when target response occurs for the entire length of a given interval (did the behavior happend for the duration of the interval?)

Front 45

Momentary Time Sampling

Back 45

Interval is scored when target response occursat the end of a given interavl (did the beahvior happen at the end of the interval?)

Front 46

Use ______ ________ to monitor behviors to be __________

Back 46

partial interval
decreased

Front 47

Use _________ __________ to monitor behaviors to be __________

Back 47

whole interval
increased

Front 48

Why Partial Interval for Decreasing Behavior and Whole Interval for Increasing?

Back 48

Partial interval method tends to over estimate beahvior (use when decreasing) whole interval underestimate behavior (use when increasing)

Front 49

Inteval Recording and Time Samples

Back 49

shorter intervals= better sample
10 sec better then 2 minutes

Front 50

Duration
2 ways to collect

Back 50

How long a behavior occurs.
1) Total Duration
2) Duration per occurence

Front 51

Latency

Back 51

Time between stimulus and start of behavior. (measure of compliance)

Front 52

Inter-Response Time

Back 52

Time between two instance of the same behavior

Long IRT = low rates of responding
Short IRT= high rates of behavior

Front 53

Trials to Criterion

Back 53

Number of times to respond before person reaches goal (criterion)
Measures the efficency of instruction
Need to record opportunities and accuracy for each trial.

Front 54

Magnitude

Back 54

Does behavior vary in strength?
Reliability an issue if beahvior not objectively defined (can be subjective)

Front 55

Topography

Back 55

What are the physical features of the behavior? (compare with a set criterion to increase reliability (ex. handwriting in a classroom, use of a ruberic)

Front 56

PLACHECK

Back 56

Planned Activity Check
-Steps
-Create schedule of activities
-Count number of people engaged n activite and compare with total number of people

Front 57

Two types of observation

Back 57

Continuous (recording for long periods of time)
Sample (discontinuous)

Front 58

Selecting Times for Observation

Back 58

Standardize times for data collection
Select times when most likely to happen/not happen

Front 59

Validity

Back 59

Did the system measure what it was supposed to measure?

Can assess with an outside source

Front 60

Reliability

Back 60

Consistency of measurement
-interobserver agreement
(can be reliable and not acccurate, low reliability = low accuracy)

Front 61

Inter-Observer Agreement

Back 61

Helps insure your observing the same behavior over time.
Should use same methods, observe same events using the same system.

Front 62

Total Count

Back 62

Percetage of agreement for frequency counts
small value/large valuate X 100
90 instances / 100 instances X 100= 90% agreement (simple agreement)- most crude way

Front 63

Mean Count per Interval

Back 63

Increases likelihood that you are documenting that observers recorded the same event
Must divide into time intervals and record instances of behaviors within the intervals
Most often used in applied settings

Front 64

Exact Count per Interval

Back 64

Most stringent method
Percentage of intervals in which observers tallied same count
Most stringent, used in research

Front 65

Trial by Trial

Back 65

used for trial type presentations.
number of trials agree/total trials X 100

Front 66

Total Duration

Back 66

For time based data, similar to simple agreement
short/long X 100
45 min/50 min X 100
crude way

Front 67

Mean Duration Per Occurance

Back 67

Calculate agreement for duration per occurance, calculate mean

Front 68

Agreement for Interval Recording

Back 68

Interval by interval (point by point)
agree/total observations X 100= percentage of agreement

Front 69

Agreement for Interval Recording- over or underestimate agreement?

Back 69

Can over estimate agreement when number of scored intervals is low.
Use when <30% of intervals are scored.

Front 70

Scored Interval Agreement

Back 70

only calculate intervals when agreement between observers occurs, if they didn't score an interval (both), your ignore it

Front 71

Interval by interval can ____ estimate agreemtn when number of scored intervals is ______

Back 71

over
high

Front 72

Unscored Interval Agreement

Back 72

Agreements on unscored/intervals where at least one is dcored as nonoccurance) X 100
Use when >70% of intervals are scored

Front 73

Reactivity

Back 73

  Person changes behavior because
they are being watched
  A type of intervention (self monitoring)
  The good and the bad (goes away the more you are in a classroom)

Front 74

Sources of Reactivity

Back 74

Obtrusive observers
Practice effects
Self-monitoring
Verbal descriptions of contngencies (if you ___ this will happen)

Front 75

Drift

Back 75

Definition of behavior changes over time

Front 76

Bias

Back 76

Observer systematically responds to some variable other than target beahvior

Front 77

Increasing Reliability

Back 77

-Research on factors
-Increase by training data collectors, simple system
-Procedural integrity
-Check for 15-20% of sessions (.80)

Front 78

Interval by interval can ____ estimate agreemtn when number of scored intervals is ______

Back 78

over
high

Front 79

Unscored Interval Agreement

Back 79

Agreements on unscored/intervals where at least one is dcored as nonoccurance) X 100
Use when >70% of intervals are scored

Front 80

Reactivity

Back 80

  Person changes behavior because
they are being watched
  A type of intervention (self monitoring)
  The good and the bad (goes away the more you are in a classroom)

Front 81

Sources of Reactivity

Back 81

Obtrusive observers
Practice effects
Self-monitoring
Verbal descriptions of contngencies (if you ___ this will happen)

Front 82

Drift

Back 82

Definition of behavior changes over time

Front 83

Bias

Back 83

Observer systematically responds to some variable other than target beahvior

Front 84

Increasing Reliability

Back 84

-Research on factors
-Increase by training data collectors, simple system
-Procedural integrity
-Check for 15-20% of sessions (.80)