Skip to contents

Per-variable taxonomy + dispatcher (R mirror of morie.variable_taxonomy)

Source: R/variable_taxonomy.R
variable_taxonomy.Rd

Classifies every column in OTIS / ARSAU datasets by Stevens-1946 level of measurement (nominal / ordinal / interval / ratio + the practical extensions boolean / date / datetime / identifier / free-text), cardinality, functional role (identifier / outcome / covariate / weight / metadata), and cross-year safety.

Details

Drives a method dispatcher (morie_recommended_summary, morie_recommended_pair_test) that picks the right statistical analysis per variable based on its measurement level.

Hard-coded invariant overrides (the data dictionary itself states these, but we encode them in code so analyses cannot accidentally violate them):

  • OTIS UniqueIndividual_ID: random per-fiscal-year reassignment -> cross_year_safe = FALSE, role = "identifier". Cross-year joins on this column are statistically meaningless.

  • ARSAU BatchFileName / Indiv_Index: per- incident identifiers -> role = "identifier".

  • ARSAU IndivInjuries_PhysicalInjuries: boolean injury outcome -> role = "outcome".

References

Stevens, S.S. (1946) "On the theory of scales of measurement." Science, 103(2684), 677-680.

Velleman, P.F. and Wilkinson, L. (1993) "Nominal, ordinal, interval, and ratio typologies are misleading." The American Statistician, 47(1), 65-72.