Optimal Effect-Size-Based Y-Axis Ranges
Overview
Calculates effect-size-informed y-axis recommendations for visualizations of independent, repeated-measures, and mixed designs. Summarizes selected error bars, estimates conditional standardized contrasts, and generalizes range-expansion rules using constants from single-factor, two-level studies of Cohen's d. Recommendations beyond that studied setting have not been empirically validated. Provides numerical results and 'ggplot2' visualizations, including summary plots and preservation of supplied plot layouts. Confidence-interval adjustment options follow the distinctions in Cousineau (2017) doi:10.5709/acp-0214-z, including difference adjustment and combined repeated-measures adjustment. Within-subject base intervals use the normalization of Cousineau (2005) doi:10.20982/tqmp.01.1.p042 with the bias correction of Morey (2008) doi:10.20982/tqmp.04.2.p061. Correlation-only intervals and intervals with both adjustments are separate options. These references support confidence-interval construction, not the range-expansion calibration.
Initial public-version numbering, based on development version V1.8.
- Rechecked release-plan Steps 1–6. Plot-only fallbacks now honor and validate a supplied measurement range. Explicit original-layout limit overrides clear previous y-scale limits before plotting, restoring censored observations while retaining the scale configuration and leaving the input plot unchanged.
- Added regressions for those plot edge cases and independent numerical checks of multilevel conditional-effect selection and three-condition Cousineau–Morey intervals. Corrected the exploratory Hedges' g warning to describe the studied calibration setting accurately.
- Added package metadata, namespace declarations, MIT licensing, and the software citation with Shuo Zang's ORCID.
- Moved the V1.8 functions into the package's
R/directory and declared runtime dependencies instead of attaching them withlibrary().
Show all changes in 1.0.0Hide the rest of the changes in 1.0.0
- Made
optimal_graph()the sole exported function. Keptoes()as an internal calculation engine, with numerical results available throughoptimal_graph(..., detail = TRUE)$oes. Updated examples and tests to follow this API. - Retained the V1.8 summary grouping and colour correction, including fill-only, shape, and linetype mappings.
- Retained the V1.7 numerical corrections for Hedges' g, Cousineau-Morey degrees of freedom, and forwarded column/factor arguments.
- Integrated the standalone numerical and plotting regressions into installed-package tests using self-contained data.
- Added a methods guide and clarified the evidence limits: *Visualizing Standardized Effect Sizes Using the Framework of Graph Perception and Data Visualization* (Shuo Zang and Denis Cousineau, 2026, unpublished manuscript) studied single-factor, two-level comparisons using Cohen's d; broader designs, other effect estimators, and generalized ER recommendations remain empirically unvalidated extensions.
- Documented that the current a constants come directly from the effect-size study and may change after further research. A separate package-methods paper is planned.
- Identified the effect-size study's associated OSF resource (Zang and Cousineau, 2025; CC BY 4.0): <https://doi.org/10.17605/OSF.IO/DRCTA>. This is a resource DOI, separate from the unpublished manuscript and software citation.
- Recorded the supplied error-bar study-data provenance (OSF DOI 10.17605/OSF.IO/DTYBF), the exact CSV reference, and its CC BY 4.0 license. Mandatory tests remain independent of external data and network access.
- Added a corrected manual error-bar Study 4B example for the project ZIP (
development/example-error-bar-study4b.R) and included error-bar data-source attribution ininst/doc/error-bar-data-source.md. - Identified Cousineau (2017), *Varieties of confidence intervals* <https://doi.org/10.5709/acp-0214-z>, as the primary reference for CI difference adjustment and combined repeated-measures adjustment. Retained Cousineau (2005) and Morey (2008) for normalization and bias correction. Clarified that
ci95_corromits sqrt(2), whileci95_dif_corrapplies both adjustments; the article supports CI construction, not OES calibration. - Added recommended graphing workflows:
oesfor y-axis ranges withsuperb::superb()for adjustable error bars, custom ggplot2 graphs withoes, or graphs generated directly byoptimal_graph(). Cited Cousineau, Goulet, and Harding (2021), <https://doi.org/10.1177/25152459211035109>. - Prepared automated checks for Linux, Windows, and macOS. Release validation is in progress; publication follows completed checks.
From NEWS.md in the release
Install
Health
- OK5 OK2026-10-21
Documentation
- Help pages
- 3 (1 internal)includes a package overview page
- Examples that run
- 100%1 run in a plain check
- Exported functions with examples
- 100%
- Documented parameters
- 100%
- Return-value docs
- 100%
- Help pages citing references
- 100%
Downloads
Dependencies
Code & Tests
People & History
- 1.0.0Latest2026-10-21 · current release ·
12 days in CRAN's queue
Listed in CRAN's incoming queue from 2026-10-09 19:57 to 2026-10-21 10:55 UTC, in 63 readings. - RR 4.6.0 released · 2026-04-24
Package metadata
- First published
- 2026-10-21
- Total releases
- 1 / 1 yrs
- License
- MIT + file LICENSE OSI
- Minimum R
- ≥ 4.1.0
- Download size
- 43 KB
Cite
Cite this package
Run in R for the authors' preferred citation:
citation("oes")The package provides a citation file that builds its entry when R runs it. Run citation() to see it.
Cite the R Observatory
For a number measured here: a download total, a coverage figure, an archival date.
From data release v2026-10-21, which the citation names so these numbers can be found later. More on citing and the projects behind them.