Textbook Notes (363,473)
Psychology (9,573)
PSYB45H3 (1,061)
Chapter 3

# Chapter 3.docx

9 Pages
49 Views

School
University of Toronto Scarborough
Department
Psychology
Course
PSYB45H3
Professor
Amanda Uliaszek
Semester
Winter

Description
Chapter 3 Using data and research methods in behavior analysis Using data to measure changes: 1) How we use data. 2) Organizing data. 3) Graphing data Using graphs and basic research methods: 1) Graphic analysis. 2) Basic research designs Advanced research designs in behavior analysis: 1) Multiple baseline designs. 2) Changing criterion and alternating treatment designs Evaluating resulting changes in behavior: 1) Dimensions of evaluation. 2) Preparing a report Tips on using graphs and research methods Using data to measure changes Using data to measure changes: Anytime applied behavior analysis is used, data must be collected and evaluated. How we use data: 1) Frequency of behavior compared to its baseline level. If intervention worked, frequency of behavior should decrease (or increase) from baseline. Intervention: A program or period of time in which action is taken to alter an existing situation, such as a target behavior Baseline: Two meanings: 1) Refers to the data collected before the intervention begins. 2) Refers to the period of time during which those data were collected. Main role of baseline data is to give a reverence point for comparison during the intervention phase. Data: Tells us the current status and history of variables. Variables: behavior, antecedents, consequences. The data we collect on these variables can clarify issues or concerns at different points in planning and conducting a program to change a target behavior (such as choosing the best techniques to apply). Organizing data Arithmetic calculations: When data varies a great deal, calculating an average or mean for a set of data smoothes out the record and gives a general level of behavior. Tables: A table is a systematic arrangement of data or other information in rows and columns for easy examination. It organizes the data visually, allowing us to see patterns and make comparisons in the data plainly and quickly. Characteristics of a table: 1) Assessment method of behavior being measured. 2) Ordered into groups. 3) Specific variables. 4) Descriptive title Graphing data Graph: A graph is a drawing that displays variations within a set of data. Typically shows how one variable changed with changes in another variable. Types of graphs: 1) Line graphs. 2) Bar graphs. 3) Cumulative graphs. Line graphs: Uses straight lines to connect successive data points that represent the intersects of plotted values for the variables scaled along the horizontal and vertical axes. Horizontal axis typically scales time or sessions (spanning baseline and intervention phases of a program). Bar graphs: Uses vertically arranged rectangles to represent data points scaled along the vertical axis. Horizontal axis is usually set conditions, and not scaled. Cumulative graphs: Measure of behavior accumulates across units scaled along the horizontal axis. This differs from normal line graphs in the successive accumulation of data points as the trial period increases. If no responses occur on the next data set, the measurement stays the same, since nothing has changed. The steeper the slope in a cumulative graph, the higher the response rate. Preparing graphs Five components of a graph: 1. Axes: 2. Axis scaling and labels 3. Data points 4. Phase lines and labels: Baseline phase vs Intervention phase 5. Caption / Title Using graphs and basic research methods Evaluating success of a program: 1) Has behavior changed? 2) Why did behavior change? Graphic analysis: Can be used to 1) Assess effectiveness of intervention. 2) Provide feedback as reinforcement for desired behavior. Judging effectiveness of a program: Assess two trends (general patterns of change in the behavior over time): 1) Whether behavior has improved from baseline to intervention. 2) Whether behavior has continued to improve across time during the intervention. Clarifying a graphic analysis Trend lines: Added to graphs to make graphic analysis clearer. A line of best fit (represents all data points within a time period). A trend line should bisect all points in half. Involves three steps: 1. Calculate the means for the baseline and intervention data you want to compare 2. Place a data point on the graph for each mean halfway across the corresponding time period 3. For each time period you’re comparing, draw a trend line. Data problems in graphic analysis Difficulties in evaluating trends: Come from data problems of three types: 1. Excessive variability 2. Decreasing baseline trend (behavioral excess) 3. Increasing baseline trend (for behavioral deficit) Excessive variability: Increasing sample size smoothes out excessive variability. Behavioral excess: Having a decreasing baseline when there’s a behavioral excess is a problem because you cannot know if behavioral change was as a result of intervention or due to natural causes. Behavioral deficit: Having an increasing baseline when there’s a behavioral deficit is a problem because you cannot know if behavioral change was as a result of intervention or due to natural causes In general: Whenever baseline data show excessive variability or an increasing or decreasing trend in relation to a behavioral goal, you should consider delaying the start of the intervention and collect additional baseline data Basic research designs Understanding why change occurred: Conduct experiment to understand why a behavioral change occurred. Research in behavioral analysis typically uses single-subject designs (or single case designs). Single-subject designs: Examines the target behavior of a person across time, while intervention is either in effect or absent. Variables: Most research includes two types of variables: 1) Independent variable. 2) Dependent variable. Independent variable: Tested for its potential or suspected influence. The presence or absence of an intervention is the independent variable. Dependent variable: Assessed to see if its value corresponds to (depends on) variations in the independent variable. Target behavior is the dependent variable. Cause and effect: Did an intervention cause the behavior to change? To determine this relationship, you must rule out the action of extraneous variables by holding them constant across your other variables. Functional relation: The behavior changes as a function of the independent variable. The AB design The AB design: A: Indicates baseline phase in which intervention was absent. B: Indicates intervention phase. AB design is useful when you need to determine the extent to which behavior changed. It is less than idea if you want to isolate the cause of the change. The reversal design, ABA design, ABAB designs The reversal design: Reversal designs have a series of phases in which an intervention is alternately absent and present. Usually with three or four phases: 3) ABA design. 4) ABAB design. Reversal designs have a distinct advantage over AB designs. They can demonstrate increases and decreases in the behavior that correspond to the presence and absence of the intervention. As a result, reversal designs are able to provide strong evidence for a functional relationship (cause and effect). Reversal designs show both that the behavior changed and why it changed. ABA design: Three phases: 1) Baseline. 2) Intervention. 3) Reversal; return to baseline. Reversal phase: Allows us to see whether the behavioral changes that occurred during intervention revert toward baseline levels when the intervention is absent ABAB design: Four phases: 1) Baseline. 2) Intervention. 3) Reversal. 4) Intervention. By reinstating intervention in the last phase, we can see whether the behavior responds again to the program’s techniques. Example of ABAB design: 1. Target behavior: reduce loud abusive statements from an individual institutionalized with mental retardation 2. Intervention applied two methods: 1) Punishment. 2) Rewards. 3. Punishment: moving the person to a corner of the room when outbursts occurred and leaving them there for 2 minutes 4. Reward: not having an outburst for certain periods of time. 5. Data shows that loud vocalizations dropped sharply when intervention was in force 6. Data shows that it returned to baseline levels during the reversal phase 7. This data suggests that her outbursts were reduced by the intervention Problems in using reversal designs Three problems: 1. Effect of the intervention may not be fully or substantially reversible (i.e. when intervention is withdrawn, the behavior may not revert towards baseline levels). Leads to difficulty interpreting whether or not it was the intervention that caused a change in behavior during the first intervention phase. 2. Must decide what conditions would constitute a reversal of the intervention. For example; Intervention involved reinforcing a behavior that was not reinforced in baseline. To arrange a reversal  terminate the reinforcement. Removing the reinforcement also removes the “contingency” … page 49 3. It may be undesirable or unethical to withdraw an intervention that appears to have produced a beneficial effect. E.g. Successfully reduced self-harm in disordered children  not ethical to return such behavior. Why doesn’t behavior revert?: Behavior may have been changed permanently by the original intervention phase (learning occurred). If you think that a target behavior will not regress to baseline, then you should not use a reversal design. Advanced research designs in behavior analysis Multiple baseline designs: More than one AB design is conducted with all baselines started at about the same time and proceeding together for a while. Each baseline continues for a different length of time before the intervention begins. Multiple baseline designs have two important characteristics: 1. No reversal phases; useful when behavior change is permanent or when withdrawing the intervention is undesirable. 2. Introduction of the intervention is staggered across the separate AB designs so that a baseline phase in at least one AB design overlaps an intervention phase in at least one other AB design Importance of overlap: Overlap between the multiple baseline designs enables comparison between target behavior in baseline with behavior in the intervention simultaneously within and across designs. 1. Allows us to see whether the behavior changed 2. Allows us to see why the behavior changed. 3. If the behavior in intervention was markedly improved over baseline in each of the multiple AB designs, we can conclude that it changed a. If a change occurred only after the intervention was introduced, we can conclude that the changes resulted from the intervention and not some other factor. Multiple baseline across behaviors design: 1. Uses separate AB designs for each of two or more different behaviors for a single individual in a particular setting. 2. Simultaneously monitor two or more different behavi
More Less

Related notes for PSYB45H3

OR

Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.