Transform AOI proportions to empirical logits — transform_gazepoint_aoi_empirical

Convert bounded AOI proportions into empirical logits for linear mixed models, growth-curve analysis, or other approximately Gaussian time-course models.

Usage

transform_gazepoint_aoi_empirical_logit(
  data,
  numerator_col = NULL,
  denominator_col = NULL,
  proportion_col = NULL,
  correction = 0.5,
  pseudo_denominator = 1,
  output_col = "aoi_empirical_logit",
  adjusted_proportion_col = "aoi_proportion_adjusted",
  raw_proportion_col = "aoi_proportion_raw",
  numerator_output_col = "aoi_numerator",
  denominator_output_col = "aoi_denominator",
  status_col = "aoi_empirical_logit_status",
  overwrite = FALSE,
  name = "gazepoint_aoi_empirical_logit"
)

Arguments

data: A data frame containing AOI proportions or AOI count data.
numerator_col: Optional numerator column, for example number of samples or fixations inside the AOI.
denominator_col: Optional denominator column, for example total valid samples or total fixations in the window.
proportion_col: Optional bounded AOI proportion column. If numerator_col and denominator_col are supplied, the raw proportion is computed from those columns. If only proportion_col is supplied, a pseudo-denominator is used and recorded in the output.
correction: Positive correction constant added to numerator and non-AOI count. The common empirical-logit correction is 0.5.
pseudo_denominator: Positive pseudo-denominator used only when proportion_col is supplied without denominator_col.
output_col: Name of the empirical-logit output column.
adjusted_proportion_col: Name of the adjusted proportion output column.
raw_proportion_col: Name of the raw proportion output column.
numerator_output_col: Name of the numerator output column used in the transformation.
denominator_output_col: Name of the denominator output column used in the transformation.
status_col: Name of the row-level transformation status column.
overwrite: Logical. If FALSE, the function errors when output columns already exist in data.
name: Character label stored in object attributes.

Value

A tibble with empirical-logit transformation columns added. The object has class gp3_aoi_empirical_logit_data.

Details

Binomial GLMMs are usually preferable when numerator and denominator counts are available. This helper is intended for sensitivity analyses, GCA-style models, and linear time-course summaries where a transformed AOI proportion is needed.