From Data to Decisions: Applying Inferential Statistics to Idiographic Longitudinal N-of-1 Data to Evaluate Intervention Effects – A Scoping Review
Abstract
Objective: This scoping review systematically maps analytical methods for evaluating intervention-related change in idiographic longitudinal N-of-1 data from AB single-case experimental designs (SCEDs) and compares their assumptions, data requirements, interpretability, and clinical applicability. Method: Following PRISMA-ScR guidance, we searched six databases from inception to May 2026 for peer-reviewed English-language studies that introduced, compared, or evaluated methods capable of determining the presence or absence of an intervention effect in idiographic or non-standardized outcomes. Methods were charted by methodological characteristics, assumptions, data requirements, clinical applicability, limitations, and software availability. Selected methods were illustrated using daily stress ratings from one clinical AB-SCED and an artificially reduced weekly dataset. Results: We identified 17 methods across 27 papers, comprising visually oriented, nonoverlap, Bayesian, simulation-based, and further approaches. No method was universally suitable. Nonoverlap indices were comparatively accessible but differed in sensitivity to trends, autocorrelation, phase length, and small samples. Bayesian and simulation-based methods offered additional inferential flexibility but required stronger assumptions or expertise. In the case illustration, methods yielded divergent classifications of improvement, deterioration, or no change; most conclusions were stable after weekly subsampling, although Bayes Factor trend evidence and statistical significance for Tau and simulation modelling analysis were attenuated. Conclusions: Method selection should be driven by the clinical question, expected effect pattern, phase length, baseline trend, autocorrelation, measurement properties, interpretability, and implementation resources. Transparent decision procedures combining visual analysis with complementary evidence on change, magnitude, and clinical relevance are needed.
Related
- Leveraging ecological momentary assessment and dynamic structural equation modeling to identify idiographic antecedents of child irritability and defiance
- A pilot evaluation of a preventive intervention to support mental health among dementia caregivers: a controlled longitudinal study
- An Idiographic Examination of Body-Focused Repetitive Behaviors Using Ecological Momentary Assessment and Network Analysis
- A longitudinal qualitative evaluation of the rollout of a complex digital intervention across healthcare organisations and the significance of implementation pace
- Calibrated large language models demonstrate a consistent advantage in a randomized longitudinal medical education intervention
BibTeX
@misc{kininger2026from,
title = {{From Data to Decisions: Applying Inferential Statistics to Idiographic Longitudinal N-of-1 Data to Evaluate Intervention Effects – A Scoping Review}},
author = {Felicitas Kininger and Simon Blackwell and Julia Glombiewski and Saskia Scholten},
howpublished = {OSF Preprints},
year = {2026},
month = oct,
doi = {10.31219/osf.io/97z23_v5},
url = {https://osf.io/97z23_v5/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - From Data to Decisions: Applying Inferential Statistics to Idiographic Longitudinal N-of-1 Data to Evaluate Intervention Effects – A Scoping Review
AU - Felicitas Kininger
AU - Simon Blackwell
AU - Julia Glombiewski
AU - Saskia Scholten
PY - 2026
DA - 2026/10/28
PB - OSF Preprints
DO - 10.31219/osf.io/97z23_v5
UR - https://osf.io/97z23_v5/
AB - Objective: This scoping review systematically maps analytical methods for evaluating intervention-related change in idiographic longitudinal N-of-1 data from AB single-case experimental designs (SCEDs) and compares their assumptions, data requirements, interpretability, and clinical applicability. Method: Following PRISMA-ScR guidance, we searched six databases from inception to May 2026 for peer-reviewed English-language studies that introduced, compared, or evaluated methods capable of determining the presence or absence of an intervention effect in idiographic or non-standardized outcomes. Methods were charted by methodological characteristics, assumptions, data requirements, clinical applicability, limitations, and software availability. Selected methods were illustrated using daily stress ratings from one clinical AB-SCED and an artificially reduced weekly dataset. Results: We identified 17 methods across 27 papers, comprising visually oriented, nonoverlap, Bayesian, simulation-based, and further approaches. No method was universally suitable. Nonoverlap indices were comparatively accessible but differed in sensitivity to trends, autocorrelation, phase length, and small samples. Bayesian and simulation-based methods offered additional inferential flexibility but required stronger assumptions or expertise. In the case illustration, methods yielded divergent classifications of improvement, deterioration, or no change; most conclusions were stable after weekly subsampling, although Bayes Factor trend evidence and statistical significance for Tau and simulation modelling analysis were attenuated. Conclusions: Method selection should be driven by the clinical question, expected effect pattern, phase length, baseline trend, autocorrelation, measurement properties, interpretability, and implementation resources. Transparent decision procedures combining visual analysis with complementary evidence on change, magnitude, and clinical relevance are needed.
ER -