Measurement
When to use whole interval recording
Want to increase a behavior (attention to task) because it's a more conservative time sampling procedure to measure behaviors to increase.
3 indicators of trustworthy measurement
VAR
V- Validity
A- Accuracy
R- Reliability
List the two types of measurement
Topography
Magnitude
Define Dependent Variable
The target behavior of interest being measured
If you're testing the effect of something on a skill, ____ is being manipulated
Independent Variable (IV)
When to use partial interval recording
For behaviors we want to DECREASE
Measurement that produces data that's applicable and specific to the target behavior and why that behavior is being measured
Validity
Human error
1. poorly created measurement systems
2. Bad observer training
3. Unintended influences on observers
Define Independent variable (IV)
The particular aspect of the environment that is manipulated assess whether it effects the target behavior
AKA: treatment; intervention; experimental variable
When to use a multiple probe design
You don't need concurrent/continuous measurement for a behavior or it is impractical
Teaching sequences (analyzing a shaping program). Probes provide baseline prior to and after each sequence being taught.
When to use momentary time sampling
when you are not able to continuously measure throughout the entire interval (as you would in whole and partial interval procedures); easily identifiable behaviors like task engagement; when measuring multiple behaviors at the same time (i.e different subjects with different behaviors + multiple settings). Planned activity checks are a variation of this.
Don't use when you care about the behavior being accounted for; low count and short duration behaviors.
Threats to validity
Indirect measures
Measuring the wrong dimension of the target behavior
Measurement artifacts (i.e data that do not provide a meaningful representation of the behavior)/ misleading picture of behavior
What defines a stable baseline?
DV values fall in a small range on the graph
NO display of upward or downward trend
Some experiments have more than 1 DV measured. Reasons for multiple DVs include:
- identifying data patterns that are treated as a control condition against evaluation and replication
- determine if there are collateral effects. (i.e is the IV effecting behaviors other than the target bx?)
- Assessing for confounding variables (is something else causing the bx change?)
In order to establish experimental control, which of the following must change reliably when the intervention is implemented and when it is absent?
Dependent Variable
A teacher recording data on the intervals of time his students are engaged in the targeted activity by measuring their behavior at the end of each interval is an example of....
Planned activity check
The degree to which repeating a measurement procedure produces the same result.
Reliability
3 parts of baseline logic
Prediction
Verification
Replication
Define replication
When you reintroduce the IV and the results are similar to previous outcomes in previous treatment conditions
Define nonparametric analysis
A type of experiment in which you present and remove the IV to see its effects on your client's DV.
Turn on/off IV
Describe difference of direct measure of behavior and product measures
Direct Measures- measuring the actual behavior as it's occurring. Real time measurement
Product Measures- measuring behavior after it has occurred by looking at the changes the behavior produced on the environment. (permanent product)
When observed values = true values
Accuracy
Define verification
When the effectiveness of the IV is proven by removing the IV, and the behavior returns to baseline values
When to use Multiple baseline design
1 IV and 2 or more DVs (i.e generalization)
target behaviors are irreversible
when it's not appropriate (ethical) or possible to remove an effective IV- use this instead of a reversal design
Define parametric analysis
Type of experiment in which you manipulate the value of the IV to see its effects on your client's DV
Use when: you want to look at the effects of the range of values of an IV on the DV
i.e a doctor prescribes varying mg of medication to see its varying effects