PSYC2900U Ch. 10 Single-Subject Research Lecture Notes PDF

11/21/2024 2 Lecture Overview Ch. 10 – Single-Subject...

11/21/2024 2 Lecture Overview Ch. 10 – Single-Subject Course reminders Research Overview of single-subject research PSYC 2900U – RESEARCH METHODS Single-subject research designs KARLA EMENO The single-subject versus group “debate” 1 1 2 3 Reminder – Assignment #3: Method (20%) 4 Upcoming Dates Due Date Extension: Submit in Canvas by 11:59 pm on Tuesday, Nov. 26th Tuesday, Nov. 26th – Assignment 3 due in Canvas by 11:59 pm You will begin Assignment 3 with a Current Study section (revised from Tuesday, Nov. 26th Assignment 2 based on TA feedback). You will then write a Method section, which will explain – in detail – how NO IN-PERSON LECTURE – Slides only will be uploaded to you would carry out your proposed study. This will include a detailed Canvas on Ch. 11 Presenting Your Research; this content description of the participants, materials, and procedure. will NOT be on the final exam This assignment will end with a Study Limitations and Future Research NO IN-PERSON TUTORIAL – Optional exam Q&A sessions via Directions section where you discuss some limitations of your proposed Google Meet instead from 1:10 to 2 pm and 4:10 to 5 pm: research study and mention some future research that could follow from https://meet.google.com/rig-rpdo-eho your proposed study. Monday, Dec. 9th, 12-2 pm (Regent Theatre) – Final Exam Detailed instructions for Assignment 3 (including the grading rubric), an example template, and an example paper have been added to Canvas. 3 4 5 6 Reminder – Final Exam What is single-subject research? Date and Time: Monday, December 9, 12-2 pm Location: Regent Theatre at 50 King St E (DTR100) A type of quantitative research design involving a small number of subjects (approximately 2-10) Exam will consist of multiple-choice and short-answer Also referred to as small-n designs (n = statistical symbol for questions. Approximately 90 marks in total on the exam. sample size) Covers Chapters 1 through 10 (Ch. 11 is NOT on the exam). Uses experimental manipulation and control to understand Exam is CLOSED-BOOK. objective behavior Worth 35% of your final grade. It is a type of experimental research with good internal validity Bring pencils, an eraser, and your student ID. It is NOT a case study, which is usually qualitative Request a deferred exam (for a valid reason) here. 5 6 1  11/21/2024 7 8 Reasons for Single-Subject Designs Reasons for Single-Subject Designs 3. It is important to study strong and consistent 1. It is important to focus intensively on the effects that have biological or social behavior of individual participants importance 2. It is important to discover causal relationships Social Validity – Treatments that have through the manipulation of an IV, careful substantial effects on important behaviors measurement of a DV, and control of and can be implemented reliably in the real- extraneous variable(s) world 7 8 9 10 Who uses single-subject research? General Features of Single-Subject Designs Early psychologists: Wilhelm Wundt – Sensation and consciousness The DV is measured repeatedly over time at regular Herman Ebbinghaus – Memory research intervals. Ivan Pavlov – Classical conditioning The study is divided into distinct phases and the B. F. Skinner – Operant conditioning participant is tested under one condition per phase. Experimental Behavior Analysts The conditions are often designated by capital Applied Behavior Analysis letters – A, B, C, and so on 9 10 General Features of Single-Subject Designs 11 General Features of Single-Subject 12 Designs (cont’d) The change from one condition to the next depends on the participant’s behavior. Steady State Strategy – The researcher waits until the participant’s behavior in one condition becomes fairly consistent before moving on to the next condition. When the DV has reached a steady state, then any change across conditions will be relatively easy to detect. 11 12 2  11/21/2024 13 14 Types of Designs Reversal Design (ABA Design) Reversal designs: ABA is the most basic single-subject research design. 1. ABA design (also ABAB, ABABA, etc.) The researcher measures the DV in 3 phases: 2. Multiple treatment reversal design 1. Baseline, before treatment is introduced (A) 3. Alternating treatments design 2. After the treatment is introduced (B) Multiple Baseline designs: 3. Return to baseline after removing the treatment (A) 1. Multiple baseline design across participants The researcher waits until the DV reaches a steady state 2. Multiple baseline design across behaviors in one phase before moving on to the next phase. 3. Multiple baseline design across settings 13 14 Reversal Design (ABA Design) 15 ABAB Reversal Design 16 15 16 17 18 Other Reversal Designs Potential Problems with Reversal Design Multiple-Treatment Reversal Design – The baseline phase is 1. If a treatment is working, it may be unethical to remove it. followed by separate phases in which different treatments Ex: A treatment that seems to reduce self-injury in a child are introduced. with an intellectual delay Participant could be returned to baseline before 2. The DV may not return to baseline when the treatment is reintroducing each treatment (ex: ABCACB design). removed. Alternating Treatments Design – Two or more treatments Ex: Positive attention for studying is removed and the are alternated relatively quickly on a regular schedule. student continues to study at an increased rate – this Can be a quick and effective way of comparing could mean that the positive attention had a lasting treatments, but only when the treatments are fast effect but could also mean that it was not the cause for acting. the effect in the first place 17 18 3  11/21/2024 19 Multiple-Baseline Design Across 20 Participants Multiple-Baseline Designs A baseline is established for each participant, and the treatment is then introduced for each one at a different time. Each participant is essentially tested in an AB design. If the DV changes when the treatment is introduced for one participant, it might be a coincidence. If the DV changes when the treatment is introduced for multiple participants, at different times, then it is unlikely to be a coincidence. 19 20 21 22 Data Analysis in Single-Subject Multiple-Baseline Designs (cont’d) Research Multiple-Baseline Design Across Behaviors: Visual inspection: Multiple baselines are established for the same Changes in the level of the DV from condition to participant but for different DVs, and the treatment is condition introduced at a different time for each DV Multiple-Baseline Design Across Settings: Trend – Gradual increase or decrease in the DV across observations Multiple baselines are established for the same participant but in different settings Latency – Time it takes DV to begin changing after a change in condition (quicker is better) 21 22 Data Analysis in Single-Subject 24 Effective Treatment Research (cont’d) Example Results can also be analyzed using statistical procedures. One option is to do what is typically done in group research (ex: ANOVA, t-test), but use means and SDs for each participant’s responses (and not averaging across participants). Percentage of non-overlapping data (PND) – The % of Ineffective Treatment responses in the treatment condition that are more Example extreme than the most extreme response in a relevant control condition (higher % = stronger treatment effect). 23 24 4  11/21/2024 Single-Subject versus Group “Debate” 25 Single-Subject versus Group “Debate” 26 (cont’d) Issues with visual inspection: Not sensitive enough to detect weak effects Concerns around external validity Can be unreliable Can the results for just a few participants generalize to Results cannot be clearly and efficiently summarized or others in the population? compared across studies Strong emphasis is placed on replication of results in Counterarguments: order to improve external validity The use of steady state strategy, and focus on strong and Strong and consistent effects, even with small samples, consistent effects, minimizes these concerns have been shown to generalize to the population (ex: Statistical analysis is increasingly common and can be used classical and operant conditioning research) in conjunction with visual inspection 25 26 27 28 Complementary Methods Converging Evidence Single-Subject Research: Every research design has strengths and weaknesses. Good for testing treatment effectiveness when the focus No single study can be considered definitive – no such is on strong, consistent, and biologically or socially thing as scientific proof, there is only scientific evidence. important effects. Converging Evidence – Examine the pattern of flaws Group Research: running through the research literature because the Good for testing treatment effectiveness at the group nature of this pattern can either support or undermine level or when effect is weak. Can look at interactions. the conclusions we wish to draw. Better option when IVs cannot be manipulated. Want to strive for progress, not perfection. 27 28 29 Questions about Chapter 10? 29 5