Chapter 5: IOA Flashcards
True Value
Does the observed value match the true value?
Believability
The extent to which the researcher convinces themself and others that the data are trustworthy and deserve interpretation.
Calibration
If errors are found as you evaluate the measurement system, you will correct and improve by making the necessary adjustments to make sure the measurement system is accurate.
Continuous Measurements
Frequency, rate, duration, IRT, latency
For low rates of behavior
Frequent Read Data Is Lovely
Direct Measurement
The observer is present and taking data, in real time, when the behavior is occurring.
Firsthand.
Discontinuous Measurements
Partial interval, whole interval, MTS.
For high rates of behavior
Please Wait a Moment
Exact Count Per Interval IOA
Number of intervals with 100% IOA / Total number of intervals x 100
Strictest IOA method.
Indirect Measurement
Observer not present when the behavior occurs. Secondhand.
Interviews, questionnaires, etc.
IOA
Inter-observer agreement: 2 or more observers are collecting data on a behavior and confirming whether their observations are congruent with one another.
Interval by Interval IOA
Number of intervals both recorded in agreement / total number of intervals x 100
Time sampling IOA
Overestimates agreement
Mean Count Per Interval IOA
Add IOA for each interval / total # of intervals x 100
Repeatability measuring IOA
Mean Duration Per Occurrence IOA
Duration IOA Bx1 + duration IOA Bx 2 + etc / # of duratino IOAs x 100
Temporal/Duration IOA
Use for duration per occurrence data.
Most precise temporal IOA.
Measurement Bias
When the data is assumed to have a certain outcome, so the observer tries to make sure their data reflects the predetermined outcome or goal.
Naive Observer
The observer, who was formally trained on the intervention, still does not fully understand why they are observing and recording data on the client’s behavior.
Observed Value
A measure produced by an observation and measurement system. Serves as the data that the researcher and others will interpret to form conclusions about an investigation.
Observer Drift
Without knowing, the observer changes the way they are implementing the system measuring the behavior.
Observer Reactivity
When the observer acts differently because they know they are being observed and results in measurement artifacts (errors in their recording and data).
Reliability
Can you repeat the measurement and get the same result?
Scored Interval IOA
Number of intervals both recorded occurrence / number of intervals 1 person recorded occurrence x 100
Use for bxs that occur at frequencies of 30% of intervals or fewer.
Time Sampling IOA
Total Count IOA
Smaller # / Larger # x 100
Repeatability measuring IOA.
Simplest repeatability measure IOA.
Overestimates the extent of actual agreement.
Total Duration IOA
Shorter duration / Longer duration x 100
Overestimates extent of actual agreement.
Trial by Trial IOA
Number of trials in 100% agreement / total number of trials x 100
Use for DIT.
True Value
Is what I’m measuring accurate?
Unscored Interval IOA
Number of intervals both recorded non-occurrence / number of intervals 1 recorded non-occurrence x 100
Use for behaviors that occur at frequencies of 70% of intervals or more.
Validity
Are you measuring the right thing?
3 Types of IOA
- Event recording/repeatability.
- Temporal/timing/duration.
- Time sampling/interval recording procedures.
Event Recording/Repeatability Measuring IOAs
- Mean count
- Total count
- Exact count
- Trial-by-trial
Make The Event Thrilling
Temporal/Timing/Duration IOAs
- Mean duration.
- Total duration
Make Time
Time Sampling/Interval Recording Procedure IOAs
- Unscored interval
- Scored interval
- Interval-by-interval
U Scored Intervals