A tutorial for software options to aid in assessing functional relations in single-case experimental designs
Unlocking Insights in Single-Case Experimental Designs with AI
This tutorial comprehensively reviews freely available web-based software for analyzing single-case experimental design (SCED) data, focusing on assessing functional relations. It outlines a structured approach to visual analysis, covering within-phase patterns, focal data features (level, trend, variability, immediacy, overlap), and consistency across multiple A-B comparisons. The tutorial emphasizes the importance of transparent reporting and provides practical guidance for applied researchers on how to use specific websites for data visualization, quantification, and effect size calculation, illustrating with real data.
Executive Impact: Quantified
Leveraging advanced AI analysis of single-case experimental designs to drive significant improvements across key performance indicators.
1.2M
Total Value Unlocked
35%
Efficiency Gains
600 hrs
Time Savings Annually
25%
Risk Reduction
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
Overview
Methodology
Key Findings
Software Applications
This section provides an overview of the challenges and importance of functional relation assessment in Single-Case Experimental Designs (SCEDs). It highlights the need for a structured approach to visual analysis to overcome interrater agreement issues and ensure transparent reporting. The tutorial specifically addresses how freely available web-based tools can facilitate this process, moving beyond simple effect size calculations to a holistic evaluation of intervention effectiveness.
The methodology emphasizes a multi-step data analytical plan, starting with within-phase pattern assessment (variability, level, trend) for each basic A-B comparison. Subsequently, focal data features such as overlap (quantified by NAP), immediacy, and projected baseline trends (conservative dual criteria, TrendMAD) are evaluated. Consistency across all basic effects is then assessed using visual aids like modified Brinley plots and two-level model outputs to determine the overall pervasiveness of the intervention effect.
The analysis of Ian's data, particularly for the 'brushing teeth' behavior, showed a Nonoverlap of All Pairs (NAP) value of 80%, indicating a moderate to strong effect for that specific comparison. However, when assessing consistency across all four behaviors, including a control behavior, the overall evidence for a consistent intervention effect based on level or trend was not strong. This demonstrates the critical role of comprehensive consistency checks in SCED analysis, even when individual basic effects show promise.
The tutorial extensively reviews and illustrates several web-based applications:
- manolov.shinyapps.io/Overlap/: For overall visual analysis, overlap (NAP), and two-level regression lines.
- manolov.shinyapps.io/TrendMASE/: For fitting various trend lines (OLS, Theil-Sen) and assessing trend accuracy.
- jepusto.shinyapps.io/scdhlm/: For between-case standardized mean differences and intraclass correlation.
- 34.251.13.245/MultiSCED/: For two-level modeling of trend, level, and immediate change, including variability quantification.
- manolov.shinyapps.io/Brinley/: For modified Brinley plots to assess consistency of effects.
- tamalkd.shinyapps.io/scda: For basic time-series plots and variability range lines.
80%
Nonoverlap of All Pairs (NAP) value for brushing teeth behavior, indicating a moderate to strong intervention effect.
Enterprise Process Flow
Assess Within-Phase Pattern (e.g., Variability)
→
Assess Focal Data Feature (e.g., Overlap via NAP)
→
Project Baseline Trend (Conservative Dual Criteria)
→
Account for Variability in Projections (TrendMAD)
→
Assess Consistency Across Basic Effects (Modified Brinley Plot)
| Feature | Visual Analysis Focus | Quantitative Analysis Capabilities |
|---|---|---|
| manolov.shinyapps.io/Overlap/ |
|
|
| manolov.shinyapps.io/TrendMASE/ |
|
|
| jepusto.shinyapps.io/scdhlm/ |
|
|
| 34.251.13.245/MultiSCED/ |
|
|
Illustrative Case: Ian's Cognitive Orientation Training
Description: The tutorial uses data from a study on 'Cognitive Orientation to daily Occupational Performance' for children with executive function deficits after acquired brain injury. Specifically, it focuses on participant Ian's multiple-baseline data across four goals/behaviors (three treated, one control).
Challenge: Analyzing ordinal Goal Attainment Scale (GAS) scores with inherent variability and identifying consistent functional relations across multiple behaviors.
Solution: Applying a structured visual analysis approach using web-based tools, emphasizing nonoverlap indices (NAP) due to ordinal data, and assessing consistency of effects across behaviors using tools like modified Brinley plots and two-level models.
Results: For Ian's 'brushing teeth' behavior, NAP indicated an 80% improvement. However, overall consistency analysis across all four behaviors (including a control that shouldn't change) revealed no strong, consistent intervention effect based on level or trend, highlighting the importance of consistency assessment in SCEDs.
Advanced ROI Calculator for SCED Implementations
Estimate the potential return on investment for integrating structured SCED analysis tools into your enterprise operations.
Estimated Annual Savings
$0
Hours Reclaimed Annually
0
Strategic Implementation Roadmap
A phased approach to integrating advanced SCED analysis into your research and operational workflows, ensuring seamless adoption and measurable outcomes.
Phase 1: Data Preparation & Initial Visualization
Format SCED data into required .txt files (score, phase, tier, id, time). Utilize 'tamalkd.shinyapps.io/scda' for initial time-series plots and within-phase variability assessment using range lines to understand data patterns.
Phase 2: Individual Basic Effect Analysis
For each A-B comparison, use 'manolov.shinyapps.io/Overlap/' to assess level, immediacy, and calculate Nonoverlap of All Pairs (NAP). Employ 'manolov.shinyapps.io/TrendMASE/' for detailed trend analysis (e.g., Theil-Sen or OLS) if trend stability is a focal feature, and 'manolov.shinyapps.io/TrendMAD' for baseline projection with variability.
Phase 3: Consistency & Functional Relation Assessment
Integrate findings across all A-B comparisons. Use 'manolov.shinyapps.io/Brinley/' for modified Brinley plots to visually inspect consistency of effects (level, trend, or variability). Apply the 'manolov.shinyapps.io/Overlap/' 'WWC Visual: Consistency' tab to see two-level regression lines. Quantify success rate based on a priori criteria (e.g., NAP > 65%) to conclude on functional relations.
Phase 4: Advanced Quantification & Reporting
For additional quantitative insights, use 'jepusto.shinyapps.io/scdhlm/' for between-case standardized mean differences or '34.251.13.245/MultiSCED/' for two-level models to quantify average effects and variability across tiers. Document all analytical decisions and software used for transparent reporting, aligning with calls for preregistration and avoiding questionable research practices.
Ready to Transform Your Data Analysis?
Connect with our experts to explore how these advanced SCED analysis techniques can be tailored for your specific enterprise needs and drive superior insights.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat