Biological Ranking and Effect Size Quantification for Treatments

Computes hierarchical agronomic rankings, percentage gains/losses relative to a control, absolute differences, and standardized effect sizes (Cohen's d) integrated with post-hoc significance groups.

Usage

agri_ranking(
  fit,
  trt = NULL,
  control = NULL,
  higher_is_better = TRUE,
  method = "tukey"
)

Arguments

fit: An object of class `"agri_fitted_model"` or `"agri_posthoc"`.
trt: Optional character string or vector specifying treatment factor(s) to rank.
control: Optional character string specifying the control / benchmark level (e.g., `"N0_Control"`, `"Control_22C"`, `"G1"`). If `NULL`, auto-detected by naming conventions.
higher_is_better: Logical, whether higher values represent superior performance (default = `TRUE`).
method: Post-hoc method to compute significance letters (default = `"tukey"`).

Value

An S3 object of class `"agri_ranking"` containing:

ranking_table: Data frame sorted by biological performance containing rank, treatment level(s), estimated mean/median, standard error, percentage change (%), absolute delta, Cohen's d, and significance group.
control_level: Name of the benchmark / control level used.
top_performer: Top-ranked treatment level and its percentage superiority.
factor_name: Treatment factor(s) evaluated.

Examples

data(tomato_rcbd, package = "agriDesignR")
fit <- fit_experiment(
  data = tomato_rcbd,
  response = "fruit_weight",
  treatment = "treatment",
  block = "block"
)
rnk <- agri_ranking(fit, trt = "treatment", control = "N0_Control")
#> Note: adjust = "tukey" was changed to "sidak"
#> because "tukey" is only appropriate for one set of pairwise comparisons
print(rnk)
#> 
#> ======================================================================
#>   agriDesignR: Biological Ranking & Effect Size Engine
#> ======================================================================
#>   Treatment Factor: treatment
#>   Benchmark/Ctrl  : N0_Control
#>   Top Performer   : N100 (+26.27% vs control)
#> 
#> 
#> Hierarchical Performance Ranking Table
#> ----------------------------------------------------------------------
#>  Rank  treatment Mean_Estimate    SE Delta_vs_Ctrl Percent_Gain Cohen_d Group
#>     1       N100       124.637 2.901        25.931        26.27    4.54     c
#>     2       N150       121.772 2.805        23.066        23.37    4.04     c
#>     3        N50       109.183 2.901        10.477        10.61    1.84     b
#>     4 N0_Control        98.706 2.805         0.000         0.00    0.00     a
#> 
#> ======================================================================
#>   Legend: Delta_vs_Ctrl = Absolute mean difference vs. benchmark
#>           Percent_Gain  = Relative percentage superiority / reduction
#>           Cohen_d       = Standardized effect size (|d| > 0.8 is strong)
#>