method.SM: SM (Selection Models) Method

In PublicationBiasBenchmark: Benchmark for Publication Bias Correction Methods

SM (Selection Models) Method

Description

Implements selection models for publication bias correction in meta-analysis. The method first fits a random effects meta-analysis model, then applies selection modeling to adjust for publication bias using the metafor package. Selection models account for the probability that studies are published based on their p-values or effect sizes. See \insertCitevevea1995general;textualPublicationBiasBenchmark for details.

Usage

## S3 method for class 'SM'
method(method_name, data, settings)

Arguments

method_name	Method name (automatically passed)
data	Data frame with yi (effect sizes) and sei (standard errors)
settings	List of method settings (see Details)

Details

The following settings are implemented

"default" or "3PSM"

3-parameter step function selection model with Maximum Likelihood estimator (method = "ML") and one step at one-sided p = 0.025 (i.e., selection for significance))

"4PSM"

4-parameter step function selection model with Maximum Likelihood estimator (method = "ML") and two steps at one-sided p = 0.025 and p = 0.50 (i.e., selection for significance and direction of the effect)

Value

Data frame with SM results

Author(s)

František Bartoš [e-mail]

References

\insertAllCited

Examples

# Generate some example data
data <- data.frame(
  yi = c(0.2, 0.3, 0.1, 0.4, 0.25),
  sei = c(0.1, 0.15, 0.08, 0.12, 0.09)
)

# Apply SM method
result <- run_method("SM", data, "3PSM")
print(result)

PublicationBiasBenchmark documentation built on March 16, 2026, 5:07 p.m.