Skip to contents

Record scientifically interpretable sensitivity bounds

Usage

masmr_bounds(
  lower,
  upper,
  meaning,
  units,
  justification,
  source,
  status,
  analysis_scale,
  calibration = NULL
)

Arguments

lower, upper

Named finite coefficient limits with identical names.

meaning, units, justification, source

Nonempty text, scalar or one per term. Record what is bounded, coefficient units, scientific rationale and source.

status

One of "evidence_informed", "elicited", or "stress_test", scalar or one per term. This records provenance, not a quality guarantee.

analysis_scale

Analysis scale, e.g. "standardized_mean_difference" or "log_odds_ratio", scalar or one per term. Must match outcome metadata.

calibration

Optional record of the calibration inputs and assumptions.

Value

A bounds object retaining limits, rationale and calibration. No default numeric bounds are chosen. External confidence limits are not hard bounds.

Examples

masmr_bounds(
  lower = c("S:M" = -.1), upper = c("S:M" = .1),
  meaning = "Assessment-by-group interaction", units = "SMD",
  justification = "Illustrative bounds on a binary interaction",
  source = "Constructed example", status = "elicited",
  analysis_scale = "standardized_mean_difference"
)
#> $table
#>   term lower upper                         meaning units
#> 1  S:M  -0.1   0.1 Assessment-by-group interaction   SMD
#>                                 justification              source   status
#> 1 Illustrative bounds on a binary interaction Constructed example elicited
#>                 analysis_scale
#> 1 standardized_mean_difference
#> 
#> $calibration
#> NULL
#> 
#> $fixed
#> [1] TRUE
#> 
#> attr(,"class")
#> [1] "masmr_bounds"