Returns the package's recommended argument names and the notation they map to in the utility-analysis literature. The glossary is intended to make the API explicit and to avoid mixing compact statistical notation with readable R argument names.
References
Sturman, M. C. (2001). Utility analysis for multiple selection devices and multiple outcomes. Journal of Human Resource Costing and Accounting, 6(2), 9-28.
Holling, H. (1998). Utility analysis of personnel selection: An overview and empirical study based on objective performance measures. Methods of Psychological Research Online, 3(1), 5-24.
Examples
# Literature: Holling (1998); Sturman (2001).
argument_glossary()
#> argument notation
#> 1 base_rate BR or phi
#> 2 selection_ratio SR
#> 3 selection_ratios SR_i
#> 4 joint_selection_ratio SR_conj
#> 5 validity r_xy
#> 6 validities r_xi,y
#> 7 predictor_cor R_XX
#> 8 R R = cor(X_1,...,X_k,Y)
#> 9 sdy SD_y
#> 10 n_selected N_s
#> 11 n_treated N_treated
#> 12 n_applicants N
#> 13 n_by_period N_t
#> 14 tenure T
#> 15 cost C
#> 16 cost_per_applicant C_A
#> 17 cost_per_stage C_j
#> 18 cost_by_period C_t
#> 19 baseline_validity r_baseline
#> 20 baseline_selection_ratio SR_baseline
#> 21 delta_z_y Delta z_y
#> 22 effect_size_d d
#> 23 range_restriction_ratio u
#> 24 variable_value V
#> 25 contribution_margin margin
#> 26 tax_rate tax
#> 27 discount_rate i
#> 28 stage_predictors stage index
#> 29 stage_selection_ratios SR_j
#> 30 stage_weights w_j
#> 31 n_sim simulation replications
#> 32 seed random seed
#> meaning
#> 1 Applicant-population probability of criterion success before selection.
#> 2 Overall proportion selected by a single cutoff or composite.
#> 3 Vector of marginal selection ratios for multiple predictors or hurdles.
#> 4 Overall conjunctive probability of passing all multiple cutoffs.
#> 5 Focal predictor-criterion validity coefficient.
#> 6 Vector of predictor-criterion correlations.
#> 7 Predictor intercorrelation matrix only.
#> 8 Full predictor-plus-criterion correlation matrix; predictors first, criterion last.
#> 9 Standard deviation of job performance in monetary or criterion units.
#> 10 Number of selected applicants/employees.
#> 11 Number of employees receiving an intervention.
#> 12 Number of applicants assessed by a selection system.
#> 13 Vector of active selected/retained employees by period.
#> 14 Expected duration of effects or tenure in periods.
#> 15 Total fixed or net cost supplied directly.
#> 16 Cost per applicant assessed by a one-stage system.
#> 17 Cost per applicant assessed at each hurdle stage.
#> 18 Cost incurred in each time period.
#> 19 Validity of the currently operating baseline system.
#> 20 Selection ratio of the baseline system; defaults to the focal selection ratio.
#> 21 Expected incremental standardized criterion gain supplied directly.
#> 22 Standardized mean difference for intervention utility.
#> 23 Ratio of unrestricted to restricted predictor standard deviations.
#> 24 Boudreau-style multiplier V; interpretation controlled by variable_value_convention.
#> 25 Direct contribution-margin multiplier overriding variable_value.
#> 26 Tax rate applied to benefits and, where appropriate, costs.
#> 27 Discount rate used for present-value calculations.
#> 28 List defining which predictors enter each hurdle stage.
#> 29 Within-stage selection ratios for a staged multiple-hurdle system.
#> 30 Optional within-stage composite weights.
#> 31 Number of Monte Carlo simulated applicants or draws.
#> 32 Optional seed for reproducible simulations.
#> note
#> 1 Use base_rate rather than BR in the public API.
#> 2 Use singular form for one decision cutoff.
#> 3 Use plural form when there is one SR per predictor or stage.
#> 4 Distinct from marginal selection_ratios; central for Thomas-Owen-Gunst tables.
#> 5 More readable than rxy, but documentation always gives the r_xy mapping.
#> 6 Used when multiple predictors enter a composite.
#> 7 Do not include the criterion in predictor_cor.
#> 8 Required by tr_multivariate() and simulation helpers.
#> 9 The package uses sdy because SDy is the conventional label in the literature.
#> 10 Avoid N when the role of the count is ambiguous.
#> 11 Used in Schmidt-Hunter-Pearlman intervention utility.
#> 12 Preferred name in v0.4.0; applicant_n is accepted as a legacy alias.
#> 13 Preferred name in v0.4.0; n_t is accepted as a legacy alias in boudreau_utility().
#> 14 Used when the same effect persists over multiple periods.
#> 15 Use more specific names when costs are per applicant, per stage, or per period.
#> 16 Preferred over CA because it is self-documenting.
#> 17 Used in multiple-hurdle simulations.
#> 18 Preferred name in v0.4.0; cost_t is accepted as a legacy alias.
#> 19 Implements the Sturman baseline critique.
#> 20 Important when baseline and focal systems have different selectivity.
#> 21 Allows direct use of outputs from simulations or external models.
#> 22 Used for training/intervention utility rather than selection validity.
#> 23 Preferred name in v0.4.0; u is accepted as a legacy alias.
#> 24 Use contribution_margin directly when V is conceptually ambiguous.
#> 25 Most transparent way to represent value retained after variable costs.
#> 26 Must be in [0, 1].
#> 27 Non-negative scalar or vector, depending on context.
#> 28 Each element is an integer vector of predictor indices.
#> 29 Product gives the expected net selection ratio in a strictly sequential design.
#> 30 Defaults to unit weights within each stage.
#> 31 Keep examples small; use larger values in applied analyses.
#> 32 Set seed for reproducible documentation and tests.
subset(argument_glossary(), argument %in% c("base_rate", "selection_ratio", "sdy"))
#> argument notation
#> 1 base_rate BR or phi
#> 2 selection_ratio SR
#> 9 sdy SD_y
#> meaning
#> 1 Applicant-population probability of criterion success before selection.
#> 2 Overall proportion selected by a single cutoff or composite.
#> 9 Standard deviation of job performance in monetary or criterion units.
#> note
#> 1 Use base_rate rather than BR in the public API.
#> 2 Use singular form for one decision cutoff.
#> 9 The package uses sdy because SDy is the conventional label in the literature.