Interobserver Agreement: the degree of consistency of data collected by two or more
Assessing Data Reliability
Two observes data and divided the small number of responses by the larger number of
responses they recorded we would be using the Session totals method of assessing IOA
Point-by-point method: we would divide the number of intervals by the total number of
intervals in which their records agreed or disagreed.
Reactivity: observing behavior can affect its performance
1. Trichotillomania is a anxiety disorder where you pull our hair.
2. Emotional eating is if your sad you eat instead of actually when your hungry.
Target Behavior: First step in beh. Modification is defining target beh which is decrease hair pulling.
Mya's target behavior is decrease emotional eating. Target behavior wont be losing weight
because that’s the end result.
How would you measure the target behaviors? 1. You would just keep a count of how many times Stan pulls his hair. This is a frequency, so we would
assess the frequency. Either we can measure it or he can do it himself which is called self-
monitoring. Or we can measure how long he pulls his hair.
2. Record what they eat or how frequently they eat. # of junk food etc.
Frequency: How many times you attend class in a week, how many times you stutter etc.
Duration: is time, how many minutes you exercised
Magnitude: intensity, how many calories you ate, to measure emotional beh. We would you
magnitude like by using a scale 0 not depressed and 10 very depressed.
Latency: time btw something happens until the behavior happens, usually between an antecedent
and an behavior. Ex: how much times your alarm goes off before you actually get up.
Quality: how good your doing something compared to the standard. Ex: grades
Trials-to-criterion: How many times or trials it took you before you successfully did something.
Percentage: how many times the beh. Occurred out of how many times the beh. Could have
occurred. Steps to do Behavior Modification
1st step--> define or whats the target beh.
When trying to modify a behavior you want to see what comes before and after this behavior that
you want to change.
2nd step-->Antecedent: before
3rd step--> Consequence: after
Direct Assessment Method: observer assesses one way glass, observer watch video tape or listens to
audio tape of person.
2 forms used In collecting data from direct observation are the A-B-C log or Summary records.
Unstructured assessment: naturalistic way of observing, observer doesn’t interfere or intervene, Structured assessment: you set up a test, you put the situation there, setting up a unnaturalistic
Drawbacks? If people know they are being watched they will act differently.
How do we time these assessments?
If we wanted to record Stan or record Mya does that mean were watching them the entire day,
There is :
Continuous Assessments: you record them the whole time
Interval Assessments: Cutting assessment down into interval maybe every hour or every 10 min etc,
and every interval you say yes or no. Whether they did it or not.
Time Sampling: Taking whole hour pick one interval out of that and doing it. So in that one hour you
may assess the first 10 min.
Drawbacks? People could lie on questionnaires. The things that happen right before the target beh. Is really Important.
Ex: Mya's emotional eating
Covert Antecedent: emotion you have right before you eat. No one can see your emotion.