Methods for TOSTt objects — TOSTt-methods • TOSTER

Methods defined for objects returned from the t_TOST and boot_t_TOST functions.

Usage

# S3 method for class 'TOSTt'
print(x, digits = 4, ...)

# S3 method for class 'TOSTt'
plot(
  x,
  type = c("simple", "cd", "c", "tnull"),
  estimates = c("raw", "SMD"),
  ci_lines,
  ci_shades,
  layout = c("stacked", "combined"),
  ...
)

describe(x, ...)

# S3 method for class 'TOSTt'
describe(x, digits = 3, ...)

Arguments

x: object of class TOSTt.
digits: Number of digits to print for p-values
...: further arguments passed through, see description of return value for details..
type: Type of plot to produce. Default is "simple" which shows point estimates with confidence intervals. Other options include consonance plots ("c"), consonance density plots ("cd"), and null distribution plots ("tnull"). Note: null distribution plots only available for estimates = "raw".
estimates: indicator of what estimates to plot; options include "raw" or "SMD". Default is is both: c("raw","SMD").
ci_lines: Confidence interval lines for plots. Default is 1-alpha*2 (e.g., alpha = 0.05 is 90%)
ci_shades: Confidence interval shades when plot type is "cd".
layout: Layout for displaying multiple estimates. Options are "stacked" (default, separate plots stacked vertically) or "combined" (single faceted plot with shared legend). Only applies when both "raw" and "SMD" are in estimates.

Value

print: Prints short summary of the tests.
plot: Returns a plot of the effects.
describe: Verbose description of results.

Examples

# example code
# Print

res1 = t_TOST(mpg ~ am, data = mtcars, eqb = 3)

res1
#> 
#> Welch Two Sample t-test
#> 
#> The equivalence test was non-significant, t(18.33) = -2.2, p = 0.98
#> The null hypothesis test was significant, t(18.33) = -3.77, p < 0.01
#> NHST: reject null significance hypothesis that the effect is equal to zero 
#> TOST: don't reject null equivalence hypothesis
#> 
#> TOST Results 
#>                 t    df p.value
#> t-test     -3.767 18.33   0.001
#> TOST Lower -2.207 18.33    0.98
#> TOST Upper -5.327 18.33 < 0.001
#> 
#> Effect Sizes 
#>                Estimate     SE                C.I. Conf. Level
#> Raw              -7.245 1.9232 [-10.5766, -3.9133]         0.9
#> Hedges's g(av)   -1.360 0.4419   [-2.032, -0.6603]         0.9
#> Note: SMD confidence intervals are an approximation. See vignette("SMD_calcs").
# Print with more digits
print(res1, digits = 6)
#> 
#> Welch Two Sample t-test
#> 
#> The equivalence test was non-significant, t(18.33) = -2.2, p = 0.98
#> The null hypothesis test was significant, t(18.33) = -3.77, p < 0.01
#> NHST: reject null significance hypothesis that the effect is equal to zero 
#> TOST: don't reject null equivalence hypothesis
#> 
#> TOST Results 
#>                   t      df p.value
#> t-test     -3.76712 18.3323   0.001
#> TOST Lower -2.20722 18.3323    0.98
#> TOST Upper -5.32702 18.3323 < 0.001
#> 
#> Effect Sizes 
#>                Estimate       SE                    C.I. Conf. Level
#> Raw            -7.24494 1.923202 [-10.576623, -3.913256]         0.9
#> Hedges's g(av) -1.35989 0.441871   [-2.031979, -0.66034]         0.9
#> Note: SMD confidence intervals are an approximation. See vignette("SMD_calcs").

# Plot with density plot - only raw values (SLOW)
#plot(res1, type = "cd", estimates = "raw")
# Plot with consonance - only raw values (SLOW)
#plot(res1, type = "c", estimates = "raw")
# Plot null distribution - only raw values
#plot(res1, type = "tnull", estimates = "raw")

# Get description of the results
describe(res1)
#> [1] "Using the Welch Two Sample t-test, a null hypothesis significance test (NHST), and a equivalence test, via two one-sided tests (TOST), were performed with an alpha-level of 0.05. These tested the null hypotheses that true mean difference is equal to 0 (NHST), and true mean difference is more extreme than -3 and 3 (TOST). The equivalence test was not significant (p = 0.98). The NHST was significant, t(18.332) = -3.77, p = 0.001 (mean difference = -7.24 90% C.I.[-10.6, -3.91]; Hedges's g(av) = -1.36 90% C.I.[-2.03, -0.66]). At the desired error rate, it can be stated that the true mean difference is not equal to 0 (i.e., no equivalence)."