Skip to contents

Lists supported measure/design/input combinations for compute_es(), with formulas, required inputs, defaults, assumptions and source references.

Usage

es_methods()

Value

A data frame with one row per supported route. Columns are:

route_id, method_version

Route identifier and computation-specification version. These identify the calculation, not the installed package version.

measure, design, input, status

Target measure, eligible design, selected input route, and implementation status.

required, optional

Inputs the user must supply, and optional route controls with their defaults. Study/effect IDs are optional for all routes.

estimate_formula, variance_formula

Point-estimate and sampling-variance formulas. Variance formulas are labelled by method; reconstructed SDs and any explicitly applied pseudo-counts are used where appropriate.

variance_methods, default_variance_method, default_correction

Accepted variance methods and defaults when the corresponding arguments are omitted. SMD/GLASS default to exact Hedges correction; other measures use none.

analysis_scale, effect_direction

Scale of yi/vi and contrast direction.

assumptions

Design and distribution requirements; these are declarations to verify in the source study, not assumptions tested by this function.

reference, source_url, source_locator

References and source locations. A missing URL means the route preserves a study's reported estimate/SE.

Details

This function returns a registry; it does not calculate or synthesize effects. Filter its rows to find the measure, design and input you intend to use, then supply that combination explicitly to compute_es. Matching numerical degrees of freedom alone does not establish design eligibility.

In formulas, n1/n2 are arm sizes, d is uncorrected standardization, yi is the analysis-scale estimate, J is the selected bias-correction factor, and R2 is full-model unadjusted R-squared. For binary comparisons a/b are group 1 events/non-events and c/d are group 2 events/non-events. Symbols named d thus have a different meaning in binary and standardized-mean formulas. See compute_es for complete definitions, boundary policies and reconstruction rules.

Normal Wald intervals and most sampling variances are approximations. Exact Hedges correction does not make the variance or confidence interval exact. Log and Fisher-z measures have analysis-scale variances; back-transformed display estimates must not be paired with those variances. Dependence between effect rows is unspecified until explicitly constructed with compute_es_covariance(). transform_es() propagates uncertainty through an explicit transformation. The registry covers the supported routes and does not claim complete coverage of Lipsey and Wilson (2001), Appendix B.

Examples

methods <- es_methods()
methods[methods$measure == "SMD",
  c("input", "required", "default_correction", "default_variance_method")]
#>            input
#> 1       means_sd
#> 2      student_t
#> 3        anova_f
#> 4        p_value
#> 5       means_se
#> 6       means_ci
#> 7 means_total_sd
#>                                                                          required
#> 1                                                        m1, sd1, n1, m2, sd2, n2
#> 2                                                                   t, df, n1, n2
#> 3                                                 f, df1, df, effect_sign, n1, n2
#> 4                                                      p, df, effect_sign, n1, n2
#> 5                                               m1, se1, n1, m2, se2, n2, se_type
#> 6 m1, lower1, upper1, n1, m2, lower2, upper2, n2, ci_level1, ci_level2, ci_method
#> 7                                        m1, m2, n1, n2, sd_total, total_sd_scope
#>   default_correction default_variance_method
#> 1       hedges_exact                      LS
#> 2       hedges_exact                      LS
#> 3       hedges_exact                      LS
#> 4       hedges_exact                      LS
#> 5       hedges_exact                      LS
#> 6       hedges_exact                      LS
#> 7       hedges_exact                      LS
methods[methods$design == "ordinary_regression",
  c("measure", "input", "analysis_scale")]
#>    measure        input          analysis_scale
#> 16    PCOR  correlation     partial_correlation
#> 17    PCOR regression_t     partial_correlation
#> 18   ZPCOR  correlation        fisher_z_partial
#> 19   ZPCOR regression_t        fisher_z_partial
#> 20   SPCOR  correlation semipartial_correlation
#> 21   SPCOR regression_t semipartial_correlation
#> 22  ZSPCOR  correlation    fisher_z_semipartial
#> 23  ZSPCOR regression_t    fisher_z_semipartial
methods[methods$input == "means_ci", c("measure", "required", "optional")]
#>    measure
#> 6      SMD
#> 10      MD
#> 13   GLASS
#> 33     ROM
#>                                                                                            required
#> 6                   m1, lower1, upper1, n1, m2, lower2, upper2, n2, ci_level1, ci_level2, ci_method
#> 10                  m1, lower1, upper1, n1, m2, lower2, upper2, n2, ci_level1, ci_level2, ci_method
#> 13 m1, lower1, upper1, n1, m2, lower2, upper2, n2, ci_level1, ci_level2, ci_method, reference_group
#> 33                  m1, lower1, upper1, n1, m2, lower2, upper2, n2, ci_level1, ci_level2, ci_method
#>                       optional
#> 6  ci_tolerance (default 1e-8)
#> 10 ci_tolerance (default 1e-8)
#> 13 ci_tolerance (default 1e-8)
#> 33 ci_tolerance (default 1e-8)