Recommend trial and participant exclusions — recommend_gazepoint

Create explicit trial-level and participant-level exclusion recommendations from Gazepoint sample-level quality information. The helper can use validity flags, gaze-coordinate missingness, pupil missingness, and optional artifact flags to produce transparent exclusion tables.

Usage

recommend_gazepoint_exclusions(
  data,
  participant_col,
  trial_col = NULL,
  condition_col = NULL,
  validity_col = NULL,
  x_col = NULL,
  y_col = NULL,
  pupil_col = NULL,
  artifact_col = NULL,
  min_trial_samples = 10L,
  max_trial_missing_prop = 0.5,
  max_trial_artifact_prop = 0.5,
  min_participant_trials = 2L,
  min_participant_valid_trials = 1L,
  max_participant_missing_prop = 0.5,
  max_participant_artifact_prop = 0.5,
  require_both_gaze_coordinates = TRUE,
  name = "gazepoint_exclusion_recommendations"
)

Arguments

data: A data frame containing sample-level or trial-level data.
participant_col: Participant identifier column.
trial_col: Optional trial identifier column.
condition_col: Optional condition column retained in summaries.
validity_col: Optional logical/numeric/character validity column.
x_col: Optional horizontal gaze coordinate column.
y_col: Optional vertical gaze coordinate column.
pupil_col: Optional pupil column.
artifact_col: Optional logical/numeric/character artifact flag column.
min_trial_samples: Minimum samples required per trial.
max_trial_missing_prop: Maximum missing/unusable sample proportion per trial.
max_trial_artifact_prop: Maximum artifact proportion per trial.
min_participant_trials: Minimum total trials required per participant.
min_participant_valid_trials: Minimum retained trials required per participant.
max_participant_missing_prop: Maximum missing/unusable sample proportion per participant.
max_participant_artifact_prop: Maximum artifact proportion per participant.
require_both_gaze_coordinates: Logical. If both gaze columns are supplied, should a sample be usable only when both coordinates are finite?
name: Character label stored in object attributes.

Value

A list with overview, trial recommendations, participant recommendations, an explicit exclusion table, and settings.

Details

This function recommends exclusions only. It does not remove rows.