Knowledge Base

Effect of Behavior Group - Frequency/Sequence Method

Mutually Exclusive Groups

In a Mutually Exclusive group, the behaviors are compared within the group. In the example below, events A, B, and C are part of the same Mutually Exclusive group. Two observations differ in that event B was scored at time 16 in Observation 1 while event C was scored in Observation 2 at that moment.

When you do not select Analyze gaps between events and run a Reliability Analysis with the Frequency/Sequence method, one disagreement is found at time 16.

See Frequency/Sequence and Gaps; Mutually Exclusive, Non-Exhaustive Behavior Groups for what happens when you include gaps between events in the analysis.

Start-Stop Groups

In a Start-Stop group, each behavior is analyzed independent from the others. The frequency/sequence method searches for matches between event A and event A, between event B and event B, and event C and event C. This means that there will be no disagreements like Event B (Obs 1) - Event C (Obs 2), that you do find with Mutually Exclusive groups. Instead you will find disagreements like Event B (Obs 1) - No event (Obs 2), and disagreements in the time events of the same type were scored in both observations.

In the example below, with similar observations as in the example above, Reliability Analysis finds two disagreements.

The first disagreement (1 in picture below) is found at time 16.00, because the time of event B in Observation 1 does not match the time of event B in Observation 2. The nearest event B in Observation 2 is from 8.00 to 12.00. This is outside the tolerance window and therefore it is a disagreement.

The second disagreement (2 in picture below) is found because event C was scored at time 16.00 in Observation 2, while no event C was scored at that moment in Observation 1.

See Frequency/Sequence and Gaps; Start-Stop Behavior Groups for what happens when you include gaps between events in the analysis.

Effect of Behavior Group - Duration/Sequence Method

One Mutually Exclusive Group

In a Mutually Exclusive group, instances of the behaviors within the group are compared. Within the group these behaviors will never overlap. Therefore, within the behavior group, the agreements and disagreements will exclude each other and the total agreement and disagreement will never exceed the observation length.

Multiple Mutually Exclusive Groups

When you have multiple Mutually Exclusive groups, behaviors from different groups may overlap. This means that also the agreements and disagreements may overlap and the total agreement and disagreement is likely to be larger than the observation length.

In the picture below events A and B form one Mutually Exclusive behavior group and events C and D another. Note that the sum of the Agreement and Disagreement is almost twice the observation length. This is because the Agreements and Disagreements in the two groups overlap.

See Duration/Sequence and Gaps; Mutually Exclusive, Non-Exhaustive Behavior Groups for what happens when you include gaps in your analysis.

Start-Stop Groups

In a Start-Stop group, behaviors may overlap within the group. Therefore agreements and disagreements may also overlap within the group. See the double disagreement at time 5.5 s in the example below.

See Duration/Sequence and Gaps; Start-Stop Behavior Groups for what happens when you include gaps in the analysis.

Reliability Analysis and Gaps

Definition of a Gap

A gap in an observation is the time interval in which no event was scored for a specific behavior group. In the following example, a gap occurs after the end of Event A and before the start of Event B. The two events form the behavior group 'Group AB'.

When Do Gaps Occur and How Are They Analyzed?

Gaps can occur:

• When suspending a live observation. See Reliability Analysis and Gaps when Suspending an Observation.
• In behavior groups in which behaviors exclude each other. Gaps in this group type can only occur if the behaviors in the group are not exhaustive, so they do not add up to 100%. The way gaps are treated differs between comparison methods. See:
  • Frequency/Sequence and Gaps; Mutually Exclusive, Non-Exhaustive Behavior Groups
  • Duration/Sequence and Gaps; Mutually Exclusive, Non-Exhaustive Behavior Groups
• In behavior groups in which behaviors can overlap. The way gaps are treated differs between comparison methods. See:
  • Frequency/Sequence and Gaps; Start-Stop Behavior Groups
  • Duration/Sequence and Gaps; Start-Stop Behavior Groups

Reliability Analysis and Gaps When Suspending an Observation

Definition of a Gap

The interval during which the observation was suspended is treated as a gap. When you score, for example, behavior A and then suspend-resume the observation, you have one occurrence of behavior A with a gap. The reliability analysis then counts two occurrences of behavior A: one before suspend, and one after resume.

Example

Consider the figure below. It shows part of the visualization of an observation which was suspended and resumed; behavior A was active before and after suspend-resume. In the reliability analysis, this is counted as two occurrences of behavior A as if A was scored again after resuming the observation.

How Is the Suspended Time Analyzed?

The suspended time is analyzed just like any other gap. This depends on the group type and the comparison method. See:

• Frequency/Sequence and Gaps; Mutually Exclusive, Non-Exhaustive Behavior Groups
• Duration/Sequence and Gaps; Mutually Exclusive, Non-Exhaustive Behavior Groups
• Frequency/Sequence and Gaps; Start-Stop Behavior Groups
• Duration/Sequence and Gaps; Start-Stop Behavior Groups