Exam 3 Review Guide for Research Methods in ABA

Goal of Applied Behavior Research

The goal of applied behavior research is to contribute to our understanding of human behavior by conducting well-controlled studies and disseminating findings.

Ways of Dissemination

Public speaking, posters, and journal publications.

Elevator Speech

Brief speech to engage and inform quickly.

Audience Impact

Audience influences presentation style and content. Match your voice volume to the size of the room and audience. Try to speak in a natural way, not too fast, not too slow, to keep the audience's attention. Keep the visuals low in word count with pictures.

What are the five journal functions?

Selection, dissemination, preservation of history, reinforcement of scientific achievement, and shaping/education.

Selection

Setting standards for research approval.

Dissemination

Sharing of the most recent findings with other professionals.

Preservation of History

Archiving past studies for future reference.

Reinforcement for Achievement

Prestige associated with published research due to strict criteria.

Shaping and Education

Continuous edits for journal quality by experts in the field.

Dependent Variable

The variable that changes as an effect of manipulation in another variable.

Independent Variable

The variable being manipulated in the study.

Name the 7 Measurement Methods

Number, rate, percent, duration (total and per occurrence), latency, trials to criterion, and magnitude

Number

a simple count of the occurrences of the behavior during a constant time period

Rate

the number of occurrences divided by the number of time units

Percent

Occurrences divided by opportunities, multiplied by 100.

Total Duration

Overall time engaged in behavior during observation.

Duration per Occurrence

The time spent on each engagement of the behavior.

Latency

The length of time from prompt to behavior engagement.

Trials to Criterion

Count of trials to reach behavior goal.

Magnitude

Strength or force of a behavior.

Whole-Interval Recording

Scores behavior if present throughout the interval. Underestimates behaviors, which makes it good for behavior deficits.

Partial-Interval Recording

Scores behavior if present at any point within the interval. Overestimates behaviors, which makes it good for behavior excess.

Momentary Time Sampling

Scores behavior at the end of the interval, which is more practical and accurate.

Reliability of Effect

the extent to which replication of a study can produce the same outcome.

Reliability of Measurement

Accuracy and consistency of measurement techniques including IOA

Interobserver Agreement (IOA)

Consistency between different observers' measurements.

Observer Drift

Gradual change in observers' definitions over time, while remaining reliable among themselves.

Observer Bias

Observers responding to a variable other than the target behavior.

Observer Influence

The awareness of an observer changes participant behavior.

Benefits of Assessing IOA

1. Determine the competence of the new observers (when IOA is low)

2. Detect observer drift (when IOA is low).

3. Increases confidence that the target behavior was clearly defined (when IOA is high)

4. Confirms that change in the data is due to change in behavior and not in data collection (when IOA is high).

Percent Occurrence/Nonoccurrence Formula

Agreements/(Agreements + Disagreements) X 100.

Internal Validity in Studies

The degree to which outside variables influence results.

External Validity in Studies

The degree to which replications of the study have a similar effect despite variation in participants, setting, or other basic procedures.

Internal vs External Validity

Inversely related. High internal validity means low external validity. Low internal validity means high external validity.

Threat to Internal Validity: History

Uncontrolled events outside the study influencing behaviors.

Threat to Internal Validity: Maturation

Natural participant changes affect study outcomes such as age, educational level, etc.

Threat to Internal Validity: Attrition

Participant dropout impacts the generalizability of results.

Threat to Internal Validity: Hawthorne Effect

Behavior changes due to awareness of observation.

Threat to Internal Validity: Testing

Repeated exposure to assessments influence participant behavior.

Threat to Internal Validity: Instrumentation

Changes in measurement tools cause inconsistent data collection.

Threat to Internal Validity: Procedural Drift

Variability in observers adherence to study procedures.

Threat to Internal Validity: Cyclical Variability

Behavior patterns linked to external cycles such as seasons or school schedules.

Threat to Internal Validity: Data Instability

Excessive variability in data hinders clear functional relationships.

Threat to Internal Validity: Multiple Treatments

Other treatments interfere with current intervention.

Ecological Validity

The extent to which a study has relevance to the natural environment and the ability to implement it in the real world. The higher the internal validity, the lower the ecological validity.

Content Validity

The degree to which baseline measures truly measure what is the focus of the treatment.

Criterion-Related Validity

How accurately a test measures the outcome it was designed to measure.

Social Validity

A measure of the acceptability and appropriateness of the target behavior and intervention procedures.

Social Significance

The degree of relative importance the behavior change will have on the participants and consumers.

Describe why subjective measures are not appropriate for science.

Subjective experiences cannot be directly observed or measured, making them unreliable. They can also be influenced by errors and personal bias.

Social Validity and Applied Dimension of ABA

It relates to the applied dimension of behavior by only studying behaviors that are socially significant to the individual and community rather than convenient for the study.

Three Types of Social Validity

Social significance of goals, social appropriateness of procedures, and social importance of the effects.

Social Significance of Goals

Three levels: broad social goals, behavioral categories, and discrete responses within categories of behavior.

Social Appropriateness of Procedures

Are the procedures acceptable for consumers? Determined through unstructured measures such as open-ended questions or structured measures such as Likert scales.

Social Importance of Effects

Impact of outcomes through comparison to normative standards, assessment of consumers, and expert evaluation.

Social Invalidity

Consumers disapprove of some components of the program and act on their disapproval through withdrawing, encouraging others to withdraw, complaining to community officials and the media, or not implementing the program's procedures after the program consultant leaves.

Future Directions for Social Validity

1. Formative evaluations of social validity throughout the intervention (as well as at the end).

2. A greater number and types of consumers should be asked to report on social validity.

Why do we replicate studies?

1. To determine internal validity, which provides strong evidence of a functional relationship.

2. To determine external validity, which shows the generality of the study.

3. To identify outliers/exceptions.

Direct Replication

the same researcher replicating the same study with new participants or the same participants under various conditions

Systematic Replication

research that is replicated with some variation to demonstrate reliability, generality, and identify exceptions

Intra-Subject Replication

Replication involving a single participant.

Inter-Subject Replication

Replication involving multiple participants.

Intra Replication and ABAB Reversal Design

Intra-subject replication shows that the individual's behavior is actually being affected by the IV by repeating the collecting baseline data, employing the intervention, removing the intervention, and implementing it again.

Inter Replication and ABAB Reversal Design

Inter-subject replication is being shown by comparing the effects of the IV across multiple participants to prove effectiveness, reliability, and generality.

Guidelines for Replication

1. Investigators, settings, materials, etc., should remain constant within and across participants.

2. The DV should be similar across participants.

3. Participant features should be similar.

4. The IV should be the same across participants unless there is a need for modification.

5. There should be minimum 3 replications

Clinical Replication

Clinical replication combines interventions into a treatment package to use for clients demonstrating similar problem behaviors and allows clinicians to implement empirically verified treatment packages.

Prediction Phase (A1)

Initial behavior patterns before intervention.

Affirmation Phase (B1)

Data showing intervention effects on behavior.

Verification Phase (A2)

Removing intervention to confirm a cause-effect relationship between IV and DV.

Replication Phase (B2)

Reapplying intervention to replicate effects of the IV on the DV.