Comprehensive effect size analysis toolkit with BlueSky R integration. Includes Cohen's d, Hedges' g, Glass' delta, eta-squared, omega-squared, Cramér's V, phi coefficient, rank-based effect sizes, and more. Essential for clinical significance assessment, meta-analyses, power analyses, and reproducible research in medical and pathological studies.
Usage
effectsize(
data,
dep,
group,
analysisType = "two_sample",
measures_cohens_d = TRUE,
measures_hedges_g = TRUE,
measures_glass_delta = FALSE,
testValue = 0,
group1Value,
group2Value,
ciWidth = 95,
interpretation = TRUE,
descriptives = TRUE,
testDetails = TRUE,
assumptionChecks = FALSE,
plotEffects = TRUE,
plotType = "forest",
plotWidth = 600,
plotHeight = 400,
measures_eta_squared = FALSE,
measures_partial_eta_squared = FALSE,
measures_omega_squared = FALSE,
measures_epsilon_squared = FALSE,
measures_cramers_v = FALSE,
measures_phi_coefficient = FALSE,
measures_cohens_w = FALSE,
measures_rank_biserial = FALSE,
measures_cliff_delta = FALSE,
measures_vargha_delaney_a = FALSE,
measures_common_language = FALSE,
measures_cohens_u3 = FALSE,
measures_probability_superiority = FALSE,
analysis_context = "ttest",
correction_method = "none",
bootstrap_ci = FALSE,
bootstrap_samples = 1000,
pooled_sd = TRUE,
welch_correction = FALSE,
paired_analysis = FALSE,
pairing_variable,
clinical_thresholds = FALSE,
small_effect_threshold = 0.2,
medium_effect_threshold = 0.5,
large_effect_threshold = 0.8,
power_analysis = FALSE,
alpha_level = 0.05,
desired_power = 0.8,
meta_analysis_format = FALSE,
study_weights,
bluesky_integration = TRUE,
comprehensive_output = FALSE,
effect_size_family = "smd",
forest_plot_advanced = FALSE,
effect_size_distribution = FALSE,
comparison_plot = FALSE,
export_format = "standard"
)Arguments
- data
The data as a data frame.
- dep
The continuous outcome variable
- group
Variable to define groups for comparison (optional for one-sample analysis)
- analysisType
Type of effect size analysis
- measures_cohens_d
Calculate Cohen's d effect size
- measures_hedges_g
Calculate Hedges' g effect size (bias-corrected)
- measures_glass_delta
Calculate Glass' delta effect size
- testValue
Reference value for one-sample analysis
- group1Value
Specific value for first group (optional)
- group2Value
Specific value for second group (optional)
- ciWidth
Confidence level for effect size confidence intervals
- interpretation
Provide Cohen's (1988) interpretation guidelines
- descriptives
Show group descriptive statistics
- testDetails
Show t-test statistics and p-values
- assumptionChecks
Perform normality and homogeneity of variance tests
- plotEffects
Create visualization of effect sizes with confidence intervals
- plotType
Type of effect size visualization
- plotWidth
Width of effect size plot
- plotHeight
Height of effect size plot
- measures_eta_squared
Calculate eta-squared effect size (ANOVA)
- measures_partial_eta_squared
Calculate partial eta-squared effect size
- measures_omega_squared
Calculate omega-squared effect size (unbiased)
- measures_epsilon_squared
Calculate epsilon-squared effect size
- measures_cramers_v
Calculate Cramér's V for categorical associations
- measures_phi_coefficient
Calculate phi coefficient for 2x2 contingency tables
- measures_cohens_w
Calculate Cohen's w for chi-square goodness of fit
- measures_rank_biserial
Calculate rank-biserial correlation for Mann-Whitney U
- measures_cliff_delta
Calculate Cliff's delta for ordinal data
- measures_vargha_delaney_a
Calculate Vargha-Delaney A measure
- measures_common_language
Calculate common language effect size (CLES)
- measures_cohens_u3
Calculate Cohen's U3 statistic
- measures_probability_superiority
Calculate probability of superiority
- analysis_context
Statistical context for effect size calculation
- correction_method
Correction method for multiple effect sizes
- bootstrap_ci
Use bootstrap methods for confidence intervals
- bootstrap_samples
Number of bootstrap samples
- pooled_sd
Use pooled standard deviation for effect size calculations
- welch_correction
Apply Welch correction for unequal variances
- paired_analysis
Analyze paired/matched data
- pairing_variable
Variable indicating pairs/matches
- clinical_thresholds
Apply clinical significance thresholds
- small_effect_threshold
Threshold for small effect size
- medium_effect_threshold
Threshold for medium effect size
- large_effect_threshold
Threshold for large effect size
- power_analysis
Include post-hoc power analysis
- alpha_level
Significance level for power calculations
- desired_power
Desired statistical power
- meta_analysis_format
Format output for meta-analysis software
- study_weights
Variable containing study weights for meta-analysis
- bluesky_integration
Use BlueSky R statistical environment features
- comprehensive_output
Include comprehensive statistical details
- effect_size_family
Focus on specific family of effect sizes
- forest_plot_advanced
Create publication-ready forest plot
- effect_size_distribution
Show sampling distribution of effect sizes
- comparison_plot
Compare multiple effect size measures
- export_format
Output formatting style
Value
A results object containing:
results$results$instructions | a html | ||||
results$results$effectSizes | Effect size measures with confidence intervals | ||||
results$results$descriptives | Group descriptive statistics | ||||
results$results$testDetails | t-test statistics and significance | ||||
results$results$assumptions | Tests of normality and variance homogeneity | ||||
results$results$comprehensiveEffectSizes | All requested effect size measures with interpretations | ||||
results$results$varianceExplained | Eta-squared, omega-squared, and related measures | ||||
results$results$associationMeasures | Cramér's V, phi coefficient, and association measures | ||||
results$results$rankBasedEffects | Non-parametric effect size measures | ||||
results$results$commonLanguageEffects | Intuitive probability-based effect sizes | ||||
results$results$powerAnalysisResults | Statistical power calculations for observed effects | ||||
results$results$metaAnalysisFormat | Formatted output for meta-analysis software | ||||
results$results$clinicalSignificance | Clinical significance evaluation using thresholds | ||||
results$results$bootstrapResults | Bootstrap-based confidence intervals for effect sizes | ||||
results$results$effectSizeGuide | a html | ||||
results$results$clinicalGuidance | a html | ||||
results$results$methodsExplanation | a html | ||||
results$results$effectPlot | Plot showing effect sizes with confidence intervals | ||||
results$results$advancedForestPlot | Publication-ready forest plot of effect sizes | ||||
results$results$effectDistributionPlot | Sampling distribution of effect sizes | ||||
results$results$comparisonPlot | Comparison of multiple effect size measures | ||||
results$results$powerCurvePlot | Power curves for different effect sizes and sample sizes |