Mean Separation from Summary Statistics — compare

Assigns significance letters to treatment means when only the means, the error mean square and its degrees of freedom are available (for example from a published table or from another program). Uses the same tests as perform_posthoc.

Usage

compare_means(
  means,
  mse,
  df_error,
  reps,
  method = "lsd",
  alpha = 0.05,
  verbose = TRUE
)

Arguments

means: Named numeric vector of treatment means.
mse: Error mean square from the ANOVA.
df_error: Error degrees of freedom.
reps: Number of replications per mean: a single number or a vector with one value per mean.
method: One of "lsd", "duncan", "tukey", "snk", "scheffe" or "bonferroni".
alpha: Significance level. Default 0.05.
verbose: Logical. If TRUE (default), prints the table.

Value

The comparison object described in perform_posthoc: groups, table, comparisons, critical_range and statistics.

Examples

m <- c(T1 = 1180, T2 = 1265, T3 = 1342, T4 = 1410, T5 = 1238)
compare_means(m, mse = 3600, df_error = 12, reps = 4, method = "duncan")
#> 
#> Duncan's MRT (alpha = 0.05, error df = 12)
#> Critical ranges: p=2 92.439, p=3 96.757, p=4 99.374, p=5 101.105
#>  level mean SE group n
#>     T4 1410 30     a 4
#>     T3 1342 30    ab 4
#>     T2 1265 30    bc 4
#>     T5 1238 30     c 4
#>     T1 1180 30     c 4
#> 
#> Means sharing a letter do not differ significantly.